Technical Field

Described embodiments relate to systems and methods for cooling computing devices. In particular, described embodiments relate to systems and methods for immersion based cooling of computing devices.

Background

A data centre usually hosts hundreds, thousands, or tens of thousands of computing devices or servers to perform computing tasks. These computing devices generate a significant amount of heat during operation. The heat generated from the computing devices must be dissipated for the computing devices to operate properly. Otherwise, the computing devices may be damaged due to the accumulated heat in the data centre. Therefore, a cooling system is required to be installed in the data centre to dissipate the heat.

Where a cooling system is used to cool computing devices, it is important that the cooling system works effectively to ensure that the computing devices can continue to operate. Faults in the cooling system may result in damage to the computing devices if they continue to operate without being cooled. Alternatively, if the computing devices are forced to cease operating, this can be costly and may have severe negative consequences where the computing devices are performing critical processing. A reliable cooling system may be necessary for proper data centre operation.

It is desired to address or ameliorate some of the disadvantages associated with such prior methods and systems, or at least to provide a useful alternative thereto.

Throughout this specification the word "comprise", or variations such as "comprises" or "comprising", will be understood to imply the inclusion of a stated element, integer or step, or group of elements, integers or steps, but not the exclusion of any other element, integer or step, or group of elements, integers or steps.

Any discussion of documents, acts, materials, devices, articles or the like which has been included in the present specification is not to be taken as an admission that any or all of these matters form part of the prior art base or were common general knowledge in the field relevant to the present disclosure as it existed before the priority date of each of the appended claims.

Summary

Some embodiments relate to a cooling system for cooling computing devices, the cooling system comprising: at least one cooling tank configured to hold a first cooling fluid and sized to at least partially immerse at least one computing device in the first cooling fluid in order for the first cooling fluid to absorb heat generated from the at least one computing device; at least one inlet pipe fluidly connected to the cooling tank to supply the cooling fluid into the cooling tank; at least one outlet pipe fluidly connected to the cooling tank to release the first cooling fluid carrying the heat absorbed from the computing device out of the cooling tank; a heat dissipation system fluidly connected to the inlet pipe to supply the first cooling fluid and fluidly connected to the outlet pipe to receive the first cooling fluid carrying the heat absorbed from the computing devices, the heat dissipation system being configured to dissipate the heat from the first cooling fluid; a pump configured to facilitate circulation of the first cooling fluid through the tank, the inlet pipe, the outlet pipe, and the heat dissipation system; and a control device, wherein the control device is configured to facilitate the cooling of the computer devices by: monitoring the temperature of the first cooling fluid within the heat dissipation system, and in response to determining that the temperature of the first cooling fluid is outside of a predetermined threshold value, adjusting the amount of heat being dissipated from the first cooling fluid.

In some embodiments, monitoring the temperature of the first cooling fluid within the heat dissipation system comprises monitoring the temperature of the first cooling fluid as it exits a heat exchanger of the heat dissipation system.

In some embodiments, monitoring the temperature of the first cooling fluid within the heat dissipation system comprises monitoring the temperature of the first cooling fluid as it exits a heat dissipater of the heat dissipation system. According to some embodiments, monitoring the temperature of the first cooling fluid within the heat dissipation system comprises calculating a temperature difference of the first cooling fluid at two locations within the heat dissipation system, and determining that the temperature of the first cooling fluid is outside of a predetermined threshold value comprises determining that the calculated temperature difference is outside of the predetermined threshold value.

In some embodiments, adjusting the amount of heat being dissipated from the first cooling fluid comprises adjusting the flow rate of a second cooling fluid through the heat dissipation system, wherein the second cooling fluid is used to absorb the heat dissipated from the first cooling fluid.

In some embodiments, the adjusting the flow rate of the second cooling fluid comprises adjusting a valve within the heat dissipation system.

According to some embodiments, adjusting the flow rate of the second cooling fluid comprises adjusting the operation of a pump.

According to some embodiments, adjusting the flow rate of the second cooling fluid comprises adjusting the operation of a fan.

In some embodiments, the second cooling fluid comprises water.

In some embodiments, the control device is further configured to: monitor the temperature of the second cooling fluid within the heat dissipation system, and in response to determining that the temperature of the second cooling fluid is outside of a predetermined threshold value, adjust the amount of heat being dissipated from the second cooling fluid.

In some embodiments, monitoring the temperature of the second cooling fluid within the heat dissipation system comprises monitoring the temperature of the second cooling fluid as it exits a heat exchanger of the heat dissipation system. In some embodiments, monitoring the temperature of the second cooling fluid within the heat dissipation system comprises monitoring the temperature of the second cooling fluid as it exits a heat dissipater of the heat dissipation system.

According to some embodiments, monitoring the temperature of the second cooling fluid within the heat dissipation system comprises calculating a temperature difference of the second cooling fluid at two locations within the heat dissipation system, and determining that the temperature of the second cooling fluid is outside of a predetermined threshold value comprises determining that the calculated temperature difference is outside of the predetermined threshold value.

According to some embodiments, adjusting the amount of heat being dissipated from the second cooling fluid comprises adjusting the flow rate of a third cooling fluid through the heat dissipation system, wherein the third cooling fluid is used to absorb the heat dissipated from the second cooling fluid.

According to some embodiments, the adjusting the flow rate of the third cooling fluid comprises adjusting the operation of a fan.

In some embodiments, the first cooling fluid comprises a dielectric fluid.

In some embodiments, the flow rate of the first cooling fluid is kept substantially stable.

According to some embodiments, the predetermined threshold is defined by a target temperature and a tolerance.

In some embodiments, the target temperature is a temperature between 30°C and 40°C.

In some embodiments, the target temperature is a temperature between 31°C and 36°C.

In some embodiments, the target temperature is a temperature between 33°C and 35°C.

In some embodiments, the tolerance is between 0.5°C and 2°C.

In some embodiments, the control device is further configured to: monitor the temperature of the second cooling fluid within the heat dissipation system, and in response to determining that the temperature of the second cooling fluid is outside of a predetermined threshold value, adjust the amount of heat being dissipated from the first cooling fluid.

According to some embodiments, the predetermined threshold is defined by a target temperature and a tolerance.

In some embodiments, the target temperature is a temperature between 20°C and 35 °C.

In some embodiments, the target temperature is a temperature between 25 °C and 30°C.

In some embodiments, the target temperature is a temperature between 27°C and 29°C.

In some embodiments, the tolerance is between 0.5°C and 2°C.

According to some embodiments, the system comprises: a plurality of cooling tanks configured to hold the first cooling fluid and sized to at least partially immerse at least one computing device in the first cooling fluid in order for the first cooling fluid to absorb heat generated from the at least one computing device; a plurality of inlet pipe fluidly connected to each cooling tank to supply the cooling fluid into the cooling tanks; and a plurality of outlet pipes fluidly connected to each cooling tank to release the first cooling fluid carrying the heat absorbed from the computing device out of the cooling tanks; wherein the heat dissipation system is fluidly connected to each inlet pipe to supply the first cooling fluid and fluidly connected to each outlet pipe to receive the first cooling fluid; and wherein the pump is configured to facilitate circulation of the first cooling fluid through the tanks, the inlet pipes, the outlet pipes, and the heat dissipation system.

In some embodiments, the monitoring of the temperature of the first cooling fluid within the heat dissipation system occurs in substantially the same location within the system as the location in which the adjustment of the amount of heat being dissipated from the first cooling fluid is conducted. Some embodiments relate to a cooling system for cooling computing devices, the cooling system comprising: at least one cooling tank configured to hold a first cooling fluid and sized to at least partially immerse at least one computing device in the first cooling fluid in order for the first cooling fluid to absorb heat generated from the at least one computing device; at least one inlet pipe fluidly connected to the cooling tank to supply the cooling fluid into the cooling tank; at least one outlet pipe fluidly connected to the cooling tank to release the first cooling fluid carrying the heat absorbed from the computing device out of the cooling tank; two heat dissipation systems fluidly connected to the inlet pipe to supply the first cooling fluid and fluidly connected to the outlet pipe to receive the first cooling fluid carrying the heat absorbed from the computing devices, the heat dissipation systems being configured to dissipate the heat from the first cooling fluid; two pumps configured to facilitate circulation of the first cooling fluid through the tank, the inlet pipe, the outlet pipe, and the heat dissipation systems , wherein each pump facilitates circulation of the first cooling fluid through a corresponding heat dissipation system; and a control device, wherein the control device is configured to facilitate the cooling of the computer devices by: monitoring the operation of each of the pumps, and in response to identifying a fault in one of the two pumps, stopping the operation of that pump and causing the other of two pumps to operate at full capacity.

In some embodiments, determining a fault in one of the two pumps comprises determining that a motor torque of a pump motor of the pump is below a predetermined threshold.

According to some embodiments, the predetermined threshold is 70% of the ordinary running torque of the pump motor.

In some embodiments, the predetermined threshold is 85% of the ordinary running torque of the pump motor.

In some embodiments, the predetermined threshold is 90% of the ordinary running torque of the pump motor. In some embodiments, each of the two pumps are configured to operate at less than their maximum required operating capacity before a fault is detected.

In some embodiments, each of the two pumps are configured to operate at 50% of their maximum required operating capacity before a fault is detected.

According to some embodiments, one of the two pumps operates at 100% of its maximum required operating capacity before a fault is detected and the other of the two pumps operates at 0% of its maximum required operating capacity before a fault is detected, and the one of the two pumps is caused to operate at 0% of its maximum required operating capacity after a fault is detected and the other of the two pumps is caused to operate at 100% of its maximum required operating capacity after a fault is detected.

In some embodiments, the heat dissipation system is configured to dissipate the heat from the first cooling fluid to a second cooling fluid.

According to some embodiments, the second cooling fluid comprises air.

According to some embodiments, the second cooling fluid comprises water.

In some embodiments, the second cooling fluid is supplied from a mains water supply.

Some embodiments further comprise a reservoir holding a reserve supply of the second cooling fluid.

Some embodiments comprise a plurality of cooling tanks configured to hold the first cooling fluid and sized to at least partially immerse at least one computing device in the first cooling fluid.

In some embodiments, the system comprises: a plurality of inlet pipe fluidly connected to each cooling tank to supply the cooling fluid into the cooling tanks; and a plurality of outlet pipes fluidly connected to each cooling tank to release the first cooling fluid carrying the heat absorbed from the computing device out of the cooling tanks; wherein the heat dissipation system is fluidly connected to each inlet pipe to supply the first cooling fluid and fluidly connected to each outlet pipe to receive the first cooling fluid; and wherein the pump is configured to facilitate circulation of the first cooling fluid through the tanks, the inlet pipes, the outlet pipes, and the heat dissipation system.

Some embodiments further comprise a first cooling fluid held within the at least one cooling tank.

Some embodiments further comprise at least one computing device located within the at least one cooling tank.

According to some embodiments, each of the two pumps are configured to operate at less than their maximum required operating capacity before a fault is detected.

Brief Description of Drawings

Figure 1 is a block diagram of a cooling system for cooling computing devices, according to some embodiments;

Figure 2 is a schematic drawing showing the cooling system of Figure 1 according to some embodiments;

Figure 3 is a flowchart showing a method performed by the system of Figure 1 according to some embodiments,

Figure 4 is a block diagram showing an alternative cooling system, according to some embodiments;

Figure 5 is a flowchart showing a method performed by the system of Figure 4 according to some embodiments; and

Figure 6 is a schematic diagram showing an embodiment of the cooling system of Figure 4 in further detail.

Description of Embodiments

Described embodiments relate to systems and methods for cooling computing devices. In particular, described embodiments relate to systems and methods for immersion based cooling of computing devices.

Large conglomerates of computing devices are used to perform such operations as hosting websites, storing data, and/or engaging in computationally complex operations such as solving proof of work equations, rendering complex images and/or running machine learning models via large neural networks. Large numbers of computing devices operating at elevated processing speeds in close proximity generate non-negligible amounts of excess heat energy. To maintain optimal performance, these computing devices must have their excess heat energy absorbed and moved away or otherwise dissipated lest the computing devices thermal throttle, leading to a reducing in processing speed, resulting in lost performance and potential failure.

Groups of computing devices may be cooled by having air moved across heat generating components, such as central processing units (CPUs), graphical processing units (GPUs), random access memory (RAM) and/or motherboard chipsets. Computing devices may also have their components cooled using a cooling liquid, such as water, routed from one heat generating component to another to form a water loop, usually attached to a reservoir and a pumping mechanism. Large numbers of computing devices may be cooled by immersing the computing devices in a non-conductive liquid, such as a synthetic hydrocarbon oil. The computing devices may be submerged in this non-conductive liquid, allowing the liquid to permeate through the chassis of the computing devices, running over the heat generating components, absorbing said heat and moving it away.

Immersion cooling may allow for computing devices to be placed more proximal to one another when compared to other methods of cooling, as the liquid may be a more effective heat sink, and may be allowed to fully contact all parts of a computing device and passively absorb heat, instead of needing to be mechanically moved across a surface such as with conventional water loops or air cooling. However, immersion cooling systems can be difficult to maintain and inefficient to operate, particularly on a large scale.

Furthermore, where an immersion cooling system is used to cool computing devices, it is important that the cooling system works effectively to ensure that the computing devices can continue to operate. Faults in the cooling system may result in damage to the computing devices if they continue to operate without being cooled. Alternatively, if the computing devices are forced to cease operating, this can be costly and may have severe negative consequences where the computing devices are performing critical processing.

Some described embodiments relate to an immersion cooling system that regulates cooling by maintaining a temperature of a coolant within a heat dissipation portion of the system, such as by monitoring the temperature of the coolant exiting from a heat exchanger. This is in contrast to some previous immersion cooling systems which operate by maintaining a temperature of coolant within the tanks holding the computing devices, such as by monitoring the temperature of the coolant exiting from such cooling tanks. Where multiple cooling tanks are used with a single heat exchanger, maintaining a temperature of a coolant within a heat dissipation portion of the system as per the described embodiments may result in increased efficiency over some previous methods.

Some described embodiments relate to an immersion cooling system with redundant heat exchange systems that can be switched on and off in case of failure. The switching may be controlled based on data relating to the operation of one or more pumps. This may be beneficial in cases where the fluid used for the immersion cooling is an oil that may degrade rubber, such that traditional pressure sensors with rubber bellows may be unsuitable. The incorporation of one or more redundant systems may be important for achieving data centre tier accreditation, in some cases.

These and other embodiments are described in further detail below.

Figure 1 shows a block diagram of a cooling system 100, according to some embodiments. System 100 may comprise one or more tanks 101 for holding computing devices 103 to be cooled. The tanks 101 may be configured to hold a first cooling fluid in which the computing devices 103 are immersed, such that the first cooling fluid absorbs the heat generated by the computing devices. System 100 may further comprise a pump system 110 for distributing the first cooling fluid through certain components of system 100.

In some embodiments, each tank 101 may be between 1000mm and 3000mm long. In some embodiments, each tank 101 may be between 400mm and 2000mm wide. In some embodiment, each tank 101 may be between 600mm and 2000mm tall. During normal operation, each tank 101 may be configured to hold between 1000 and 2000L of cooling fluid. Each tank 101 may be configured to hold around I400L of cooling fluid, for example.

System 100 may further comprise a heat dissipation system 160. Heat dissipation system 160 may comprise a heat dissipater 125 for dissipating heat absorbed from the computing devices 103 via the first cooling fluid. For example, in some embodiments heat dissipater 125 may cause air to be passed in close proximity to the first cooling fluid in order to cause the heat from the first cooling fluid to be dissipated into the air. In such embodiments, the air may be considered a second cooling fluid. In some embodiments, heat dissipater 125 may use evaporative cooling methods to cool the air.

In some embodiments, heat dissipation system 160 may optionally comprise a heat exchanger 115 for dissipating heat from a first cooling fluid to a second cooling fluid. If heat exchanger 115 is used in the heat dissipation system 160, heat dissipater 125 may instead be configured to dissipate heat from the second cooling fluid. Specifically, in such embodiments the first cooling fluid may pass through the heat exchanger 115 in close proximity to a second cooling fluid, to cause the heat absorbed by the first cooling fluid to be dissipated into the second cooling fluid. The heat absorbed by the second cooling fluid may then be dissipated by causing the second cooling fluid to flow through heat dissipater 125, which may cause air to be passed in close proximity to the second cooling fluid in order to cause the heat from the second cooling fluid to be dissipated into the air. In such embodiments, the air may be considered a third cooling fluid. Where no heat exchanger 115 is included in system 100, air passing through heat dissipater 125 may be considered a second cooling fluid, as described above. The second cooling fluid may be water, in some embodiments, and may be supplied to system 100 by a connection to a mains water supply.

In some embodiments, system 100 may optionally comprise a first fluid reservoir 120 for holding additional supply of the first cooling fluid to supply to tanks 101 via pump system 110. In some embodiments, system 100 may optionally comprise a second fluid reservoir 140 for holding additional supply of the second cooling fluid to supply to heat dissipater 125. Where the second cooling fluid is water, the second fluid reservoir 140 may be used to hold a reserve supply of water in case the mains water supply is disconnected or disrupted. In some embodiments, the main water supply may be considered to be the second fluid reservoir 140.

System 100 may further comprise a control device 130 for controlling the components of system 100. This may include controlling the operation of one or more of the pump system 110, fluid pumps 112/114, heat exchanger 115, heat dissipater 125, or fluid control mechanism 127, for example. The system 100 may be used to cool computing devices 103 during operation. The tanks 101 may be in fluidic connection with pump system 110, heat dissipation system 160, and optionally fluid reservoir 120 by one or more fluid conduits 150, 152, 154 and 156. Tanks 101 may comprise one or more computing device racks which may be configured to hold, support or house one or more computing devices 103. Tanks 101 may be positioned into racks and/or rows of tanks that share common coolant conduits (as shown in Figure 2) that connect each tank 101 to each other tank 101 and to the rest of system 100. Neighbouring tanks 101 may also be fluidly connected to each other by one or more balance conduits and/or one or more overflow conduits.

The pump system 110 may be in fluidic communication with heat dissipation system 160, first fluid reservoir 120 and one or more tanks 101 by one or more coolant conduits 150, 152, 154 and 156. Pump system 110 may be configured to pump the first cooling fluid throughout the system 100 to collect excess heat energy from computing devices 103 in tanks 101 and to deliver the fluid to heat dissipation system 160. The first cooling fluid may be cooled by heat dissipation system 160, then pumped back through tanks 101 to absorb further heat from the computing devices 103.

In some embodiments, pump system 110 may comprise one or more fluid pumps 112, 114. Pump system 110 may comprise multiple fluid pumps depending on the volume of first cooling fluid contained in system 100, and/or the number of tanks 101 connected to system 100. In some embodiments, fluid pump 112 may be a primary or first coolant pump and fluid pump 114 may be a secondary, back-up or second coolant pump. Fluid pump 114 may be kept at an operational set-point, such as a set idle rate and/or power usage, for example such as 50% of its full power, and be caused to increase or decrease its power output upon a trigger event being detected, in some embodiments. A trigger event may comprise a power spike, a malfunction event, a reduction in coolant velocity, an increase in coolant velocity and/or any other trigger that may represent anomalous and/or non-standard behaviour of the system 100.

The heat dissipation system 160 may be in fluidic connection with pump system 110, fluid reservoir 120, and/or tanks 101 by one or more coolant pipes 150, 152, 154 and 156. Heat dissipation system 160 may comprise a heat dissipater 125, and optionally a heat exchanger 115. Heat exchanger 115 may be in fluidic connection with heat dissipater 125 by one or more coolant pipes 158, 159. Where heat dissipation system 160 includes a heat exchanger 115, heat exchanger 115 may provide an interface for heat to be dissipated between the first cooling fluid and the second cooling fluid, without allowing any direct contact between the first cooling fluid and the second cooling fluid. Where the first cooling fluid is an oil and the second cooling fluid is water, heat exchanger 115 may provide an oil to water interface, wherein the oil flowing between the tanks 101, pump system 110 and heat exchanger 115 may transfer collected heat energy to a volume of water. Heat exchanger 115 may exchange heat from the first cooling fluid to the second cooling fluid by bringing a conduit for the first cooling fluid into close proximity with a conduit for the second cooling fluid, to allow the heat energy within the first cooling fluid to transfer to the second cooling fluid.

Where heat dissipation system 160 includes a heat exchanger 115, heat dissipater 125 may be configured to receive the second cooling fluid from the heat exchanger 115. The second cooling fluid may be water in some embodiments. The second cooling fluid may have absorbed heat from the first cooling fluid, and heat dissipater 125 may be configured to dissipate the heat from the second cooling fluid to the surrounding environment and/or atmosphere. For example, in some embodiments, heat dissipater 125 may be an adiabatic cooling system. The adiabatic cooling system may comprise adiabatic pads that are wet with a fluid such as water which is caused to evaporate in a stream of warm, dry (low humidity) air acting as a third cooling fluid. In the process of going from a liquid to a gas, the evaporating fluid on the adiabatic pads simultaneously humidifies and cools the air stream, which can be used to cool the second cooling fluid. In some embodiments, heat dissipater 125 may comprise one or more of a water tower, natural water source, and/or geothermal conduit system, for example, which may be configured to transfer heat from the system 100 into the surrounding environment to cool the system coolant back down, before being fed back into the system 100.

Alternatively, where no heat exchanger 115 exists in system 100, heat dissipater 125 may be configured to receive the first cooling fluid, which may be oil in some embodiments, and to dissipate the heat from the first cooling fluid to the surrounding environment and/or atmosphere via a stream of air. In such embodiments, the stream of air may be considered a second cooling fluid.

In some embodiments, heat energy collected by the first and/or second cooling fluid may be reclaimed and used for one or more additional purposes in the data centre, such as preheating tap water, processing water, suppling air, and/or to heat other areas of the data centre, for example. Heat dissipation system 160 may further comprise one or more fluid control mechanisms 127 for controlling the movement of the first and/or second cooling fluids through the heat exchanger 115 and/or heat dissipater 125, or for moving the second cooling fluid between the heat exchanger 115 and the heat dissipater 125. Heat dissipation system 160 may also comprise one or more temperature sensors 117 for measuring the temperature of the first or second cooling fluid as it enters, exits, or passes through the heat exchanger 115 or the heat dissipater 125 . For example, temperature sensor 117 may be positioned to measure one or more of the temperature of the first cooling fluid entering the heat exchanger 115, the temperature of the first cooling fluid exiting the heat exchanger 115, the temperature of the second cooling fluid entering the heat exchanger 115, the temperature of the second cooling fluid exiting the heat exchanger 115, the temperature of the first cooling fluid entering the heat dissipater 125, the temperature of the first cooling fluid exiting the heat dissipater 125, the temperature of the second cooling fluid entering the heat dissipater 125, and/or the temperature of the second cooling fluid exiting the heat dissipater 125.

The fluid control mechanism 127 may be configured to control movement of the second cooling fluid, in some embodiments. This may include controlling movement of the second cooling fluid through the heat dissipater 125 and/or controlling movement of the second cooling fluid between heat dissipater 125 and heat exchanger 115, for example. In some embodiments, fluid control mechanism 127 may be located between heat exchanger 115 and heat dissipater 125. In some embodiments, components of fluid control mechanism 127 may exist elsewhere in system 100. Fluid control mechanism 127 may comprise at least one of a valve or a pump in some embodiments. Where the second cooling fluid is air, the fluid control mechanism 127 may comprise one or more fans.

First fluid reservoir 120 may be in fluidic connection with tanks 101, pump system 110, and heat dissipation system 160 by one or more coolant conduits 150, 152, 154 and 156. Fluid reservoir 120 may be configured to collect and/or store some or all of the volume of the first cooling fluid within the system 100. The first cooling fluid may be caused to be conveyed to the fluid reservoir 120 by pump system 110 when the system 100 or components of system 100 such as pump system 110, tanks 101, computing devices 103, heat exchanger 115 and/or heat dissipater 125 require maintenance, replacement and/or reconfiguration. Fluid reservoir 120 may be mechanically isolated from the rest of system 100 until the first cooling fluid is required to be moved to and stored in fluid reservoir 120. In some embodiments, fluid reservoir 120 may be a moveable container that may be connected or removed from the system 100 as needed.

Control device 130 may comprise one or more computing devices in communication with one or more electrical control systems of system 100. In some embodiments, control device 130 may further be in communication with one or more computing devices 103. Control device 130 may be configured to receive data from one or more sensors located in system 100, and to control one or more devices of system 100, such as the pump system 110, fluid control mechanism 127, heat exchanger 115, fluid reservoir 120 and/or heat dissipater 125. Control device 130 may be configured to monitor and/or send instructions to these and other components of system 100.

In some embodiments, control device 130 may be configured to monitor the temperature of the first or second cooling fluid at one or more locations in system 100. Control device 130 may further be configured to adjust the operation of one or more components of system 100 in order to adjust the temperature of the first or second cooling fluid at one or more locations in system 100. For example, according to some embodiments, control device 130 may receive a temperature measurement from a temperature sensor relating to a temperature of the first or second cooling fluid at a location within system 100. Control device 130 may compare the measured temperature to a predetermined temperature threshold, which may be a threshold temperature range in some embodiments. According to some embodiments, the predetermined temperature threshold may be a target temperature and a threshold amount of tolerance, such as +/- 0.5°C, for example. If the measured temperature is outside of the required threshold, control device 130 may alter the operation of one or more components in system 100 to cause the temperature to adjust to a value within the threshold.

According to some embodiments, the monitoring of the temperature by control device 130 of the first or second cooling fluid within system 100 may occur in substantially the same location within system 100 as the location in which the adjustment of the temperature of the first or second cooling fluid is being conducted. In other words, if the control device 130 is configured to adjust the operation of the heat exchanger 115 to cause the temperature of the first cooling fluid passing through heat exchanger 115 to be adjusted, then the temperature being monitored may be a temperature of the first cooling fluid at or near heat exchanger 115. In some embodiments, control device 130 may receive a temperature measurement from temperature sensor 117, being a temperature measurement corresponding to a temperature of the first or second fluid within the heat dissipation system 160. For example, the temperature measurement may correspond to the temperature of the first fluid as it exits the heat exchanger 115, once heat from the first cooling fluid has been dissipated into the second cooling fluid. In some embodiments, the temperature measurement may correspond to the temperature of the first cooling fluid as it enters heat exchanger 115 before dissipating heat into the second cooling fluid. In some embodiments, the temperature measurement may correspond to the temperature of the second cooling fluid as it enters heat exchanger 115 before absorbing heat from the first cooling fluid. In some embodiments, the temperature measurement may correspond to the temperature of the second cooling fluid as it exits heat exchanger 115 after absorbing heat from the first cooling fluid. In some embodiments, the temperature measurement may correspond to the temperature of the first or second cooling fluid as it enters heat dissipater 125 before dissipating heat. In some embodiments, the temperature measurement may correspond to the temperature of the first or second cooling fluid as it exits heat dissipater 125 after dissipating heat.

According to some embodiments, control device 130 may calculate a temperature difference for the first and/or second fluid based on the received temperature measurements. For example, control device 130 may determine the difference in temperature of the first cooling fluid as it enters heat exchanger 115 before dissipating heat into the second cooling fluid and the temperature of the first cooling fluid as it exits heat exchanger 115 after dissipating heat into the second cooling fluid. In some embodiments, control device 130 may determine the difference in temperature of the second cooling fluid as it enters heat exchanger 115 before absorbing heat from the first cooling fluid and the temperature of the second cooling fluid as it exits heat exchanger 115 after absorbing heat from the first cooling fluid. In some embodiments, control device 130 may determine the difference in temperature between the first or second cooling fluid as it enters heat dissipater 125 before dissipating heat and the temperature of the first or second cooling fluid as it exits heat dissipater 125 after dissipating heat.

Control device 130 may compare the measured temperature or the calculated temperature difference to a predetermined threshold, which may be a threshold temperature range in some embodiments. The threshold temperature range may be defined by a target temperature and an allowable tolerance in some embodiments. According to some embodiments, the target temperature may be between 30°C and 40°C in some embodiments. In some embodiments, the target temperature may be between 31°C and 37°C. In some embodiments, the target temperature may be between 33°C and 35°C.

Control device 130 may be configured to determine whether the measured temperature or the calculated temperature difference is within a tolerance from the target temperature . The tolerance may be between 0°C and 2°C in some embodiments. For example, control device 130 may be configured to determine whether the measured temperature is within 2°C of the target temperature. In some embodiments, control device 130 may be configured to determine whether the measured temperature is within 1°C of the target temperature. In some embodiments, control device 130 may be configured to determine whether the measured temperature is within 0.5°C of the target temperature.

In some embodiments, control device 130 may additionally or alternatively receive a temperature measurement from temperature sensor 117 of the second cooling fluid as it exits the heat exchanger 115, once heat from the first cooling fluid has been dissipated into the second cooling fluid. Control device 130 may compare the measured temperature to a predetermined temperature threshold, which may be a threshold temperature range in some embodiments. The threshold temperature range may be defined by a target temperature and an allowable tolerance in some embodiments. For example, the target temperature may be between 25 °C and 35 °C in some embodiments. In some embodiments, the target temperature may be between 25°C and 31°C. In some embodiments, the target temperature may be between 27°C and 29°C.

Control device 130 may be configured to determine whether the measured temperature is within a tolerance from the target temperature. The tolerance may be between 0°C and 2°C in some embodiments. For example, control device 130 may be configured to determine whether the measured temperature is within 2°C of the target temperature. In some embodiments, control device 130 may be configured to determine whether the measured temperature is within 1°C of the target temperature. In some embodiments, control device 130 may be configured to determine whether the measured temperature is within 0.5°C of the target temperature.

If the measured temperature or the calculated temperature difference is outside of the required threshold, control device 130 may alter the operation of the fluid control mechanism 127 to cause the temperature to adjust toward a value within the threshold. For example, control device 130 may alter the operation of the fluid control mechanism 127 to cause more heat to be dissipated from the first cooling fluid to the second cooling fluid. In some cases, control device 130 may instead alter the operation of the fluid control mechanism 127 to cause less heat to be dissipated from the first cooling fluid to the second cooling fluid. In some cases, control device 130 may alter the operation ofthe fluid control mechanism 127 to cause more or less heat to be dissipated from the second cooling fluid.

In some embodiments, control device 130 may alter the operation of the fluid control mechanism 127 to alter the flow rate of the second cooling fluid. This may be by increasing or decreasing the flow rate of second cooling fluid. In some cases, control device 130 may instead alter the operation of the fluid control mechanism 127 to alter the flow rate of air flowing through the heat dissipater 125, which may be acting as a second or third cooling fluid. According to some embodiments, fluid control mechanism 127 may comprise a pump, and changing the flow rate of the second cooling fluid may be achieved by altering the operation of the pump . According to some embodiments, fluid control mechanism 127 may comprise a valve, and changing the flow rate of the second cooling fluid may be achieved by at least partially opening or closing the valve. According to some embodiments, fluid control mechanism 127 may comprise a fan, and changing the flow rate of the air flowing through heat dissipater 125 and acting as a second or third cooling fluid may be achieved by altering the operation of the fan. According to some embodiments, control device 130 may alter the operation ofthe fluid control mechanism 127 to cause the temperature to adjust toward a value within the threshold without changing the flow rate of the first cooling fluid through system 100.

As shown in F igure 1 , pump system 110 may be fluidly connected to tanks 101 by coolant outlet conduit 150, such that the first cooling fluid may be pumped from or otherwise removed from tanks 101. In some embodiments, pump system 110 may also be in fluidic connection with fluid reservoir 120 by drain and fill pipes 154. Tanks 101 may be fluidly connected to heat dissipation system 160 via coolant inlet conduit 156. Heat dissipation system 160 may be fluidly connected with pump system 110 by coolant return conduit 152. Within heat dissipation system 160, heat exchanger 115 may be fluidly connected to heat dissipater 125 by one or more coolant dissipation conduits 158 and 159. Conduit

158 may supply the second cooling fluid to the heat exchanger 115 by cooling and returning second cooling fluid that has previously passed to heat dissipater 125. Conduit

159 may return the heated second cooling fluid back to heat dissipater 125 for cooling. Figure 2 shows a further schematic diagram of cooling system 100. As shown in Figure 2, the illustrated cooling system 100 includes 12 cooling tanks 101 arranged in two rows of six. As another example, the cooling system 100 may include 6 cooling tanks 101 in two rows of three . It should be noted that the illustrated cooling system 100 is an example only, and cooling system 100 can include more or less cooling tank(s) 101 without departing from the scope of the present disclosure. Further, the cooling tanks 101 can be arranged in one row or more than two rows without departing from the scope of the described embodiments. As another example, a cooling system 100 as described with reference to Figure 2 may be used as a sub cooling system, and another cooling system 100 may be used as another sub cooling system. The two sub cooling systems 100 can be fluidly connected together to form a cooling system. Such a cooling system including more than one sub cooling systems 100 is also described in the present disclosure. Further, a pipe in the present disclosure can be a straight pipe, a bent pipe, a curved pipe, or a combination of pipes in different shapes. A pipe can also include one or more segments that are fluidly connected. The one or more segments of a pipe can extend towards the same direction or different directions. Further, the reference to the segments of a pipe is not to define the structure of the pipe, but to indicate different portions of the pipe for easy description.

For ease of description, the multiple cooling tanks 101 are denoted as 101. Each of the cooling tanks 101 is configured to accommodate a first cooling fluid and sized to at least partially immerse the computing devices 103 (not shown in Figure 2) in the first cooling fluid so the first cooling fluid can absorb heat generated from the computing devices 103 so as to cool the computing devices 103. During operation of the cooling system 100 and the data centre, one or more computing devices 103 or servers are placed in each of the cooling tanks 101, and the heat generated from the computing devices 103 is absorbed by the first cooling fluid in the cooling tanks 101 to control the temperature of the computing devices 103. As a result, the cooling fluid around the computing devices 103 heats up and its temperature becomes higher.

The first cooling fluid may be a liquid coolant, and can be for example a type of dielectric fluid. A dielectric fluid is a non-conductive fluid that has a very high resistance to electrical breakdown event at high voltages. Electrical breakdown or dielectric breakdown is when an insulating (i.e. non-conductive) material becomes an electrical conductor. The present system may use such dielectric fluids as oils, for example synthetic oil, mineral oil or bioorganic oil, or an engineered fluid, such as 3M’s Novec or Fluorinert lines. In some embodiments, the system 100 may use a synthetic hydrocarbon oil, as synthetic hydrocarbon oils repel water and other foreign bodies . In some other embodiments, the oil may be an ester oil. A synthetic hydrocarbon oil may be preferable in some embodiments, as ester oils may attract water, dirt and other particles which may negatively impact the operation of the system in some cases.

The cooling system 100 also comprises a set of inlet pipes 156. The set of inlet pipes 156 are fluidly connected to the multiple cooling tanks 101 to supply the first cooling fluid into the multiple cooling tanks 101.

The cooling system 100 also comprises a set of outlet pipes 150. The set of outlet pipes 150 are fluidly connected to the multiple cooling tanks 101 to convey the first cooling fluid carrying the heat absorbed from the computing devices 103 out of the multiple cooling tanks 101. The first cooling fluid carrying the heat is conveyed out of the cooling tanks 101 from the bottoms of the cooling tanks 101 via the set of outlet pipes 150.

The heat dissipation systems 160 fluidly connects to each of the set of inlet pipes 156 directly or indirectly to supply the first cooling fluid into the set of inlet pipes 156. The heat dissipation systems 160 also fluidly connects to each of the set of outlet pipes 150 directly or indirectly to receive from the set of outlet pipes 150 the first cooling fluid carrying the heat absorbed from the computing devices 103. The heat dissipation systems 160 are configured to dissipate the heat from the first cooling fluid carrying the heat. Therefore, the temperature of the first cooling fluid is reduced and the first cooling fluid is supplied into the set of inlet pipes 156 and in turn the multiple cooling tanks 101 to cool down the computing devices 103 immersed in the first cooling fluid in the multiple cooling tanks 101.

The pump system 110 fluidly connects to each of the set of outlet pipes 150 directly or indirectly. The pump system 110 is configured to facilitate circulation of the first cooling fluid in the multiple cooling tanks 101, the set of inlet pipes 156, the set of outlet pipes 150, and the heat dissipation system 160. In the cooling system 100, the heat dissipation system 160 fluidly connects to the multiple cooling tanks 101 via the set of inlet pipes 156 and the set of outlet pipes 150. Further, the pump system 110 fluidly connects to the multiple cooling tanks 101 via the set of outlet pipes 150. Such a structure makes it unnecessary for the multiple cooling tanks 101 to have their individual heat dissipation systems and their individual coolant pumps to dissipate the heat and circulate the first cooling fluid because the heat dissipation system 160 and the pump system 110 are shared by the multiple cooling tanks 101. Therefore, the cooling system 100 allows a scalable deployment of the data centre. While two heat dissipation systems 160 and two fluid pumps 112/114 are shown in the illustrated embodiment, each heat dissipation system 160 and two fluid pump 112/114 can service a plurality of tanks 101. In some embodiments, a single heat dissipation system 160 and/or a single fluid pump 112/114 may be used.

The cooling system 100 may also comprise a second cooling fluid supply pipe 158 (not shown) fluidly connected to the heat exchanger 115 to supply a second cooling fluid (for example, cool water) into the heat exchanger 115 in order for the heat exchanger 115 of the heat dissipation system 160 to dissipate the heat into the second cooling fluid. The cooling system 100 also comprises a second cooling fluid release pipe 159 (not shown) fluidly connected to the heat exchanger 115 to release from the heat exchanger 115 the second cooling fluid with the heat (i.e., hot water).

Figure 3 shows a flowchart illustrating a method 300 that may be carried out by control device 130 to facilitate cooling of computing devices 103 located in tanks 101.

At step 310, control device 130 receives a temperature measurement relating to a temperature measured at a location within system 100. The temperature measurement may be received from temperature sensor 117, for example. According to some embodiments, the temperature measurement may correspond to a temperature measured in or near heat dissipation system 160, which may be in or near heat exchanger 115 in some embodiments. For example, the temperature measured may be the temperature of the first cooling fluid as it exits heat exchanger 115 after dissipating heat into the second cooling fluid. In some embodiments, the temperature measured may be the temperature of the first cooling fluid as it enters heat exchanger 115 before dissipating heat into the second cooling fluid. In some embodiments, the temperature measured may be the temperature of the second cooling fluid as it enters heat exchanger 115 before absorbing heat from the first cooling fluid. In some embodiments, the temperature measured may be the temperature of the second cooling fluid as it exits heat exchanger 115 after absorbing heat from the first cooling fluid. In some embodiments, the temperature measured may be the temperature of the first or second cooling fluid as it enters heat dissipater 125 before dissipating heat. In some embodiments, the temperature measured may be the temperature of the first or second cooling fluid as it exits heat dissipater 125 after dissipating heat.

At step 320, control device 130 compares the received measured temperature to a predetermined temperature threshold, which may be a temperature that corresponds to a desired operation of system 100. The temperature threshold may be retrieved from a memory storage location accessible to control device 130, in some embodiments. According to some embodiments, the temperature threshold may be a range of values within a predetermined tolerance of a target temperature value. In some embodiments, the target temperature value may be between 25°C and 35°C. In some embodiments, the target temperature value may be between 25°C and 31 °C. In some embodiments, the target temperature value may be between 27°C and 29°C. The tolerance may be between 0°C and 2°C in some embodiments. According to some embodiments, the predetermined tolerance may be +/-2°C. According to some embodiments, the predetermined tolerance may be +/-1°C. According to some embodiments, the predetermined tolerance may be +/-0.5°C. The temperature threshold may be a range of values defined by both the target temperature value and the tolerance, such that the minimum value of the range is the target temperature value minus the tolerance and the maximum value of the range is the target temperature value plus the tolerance, for example.

At step 330, control device 130 determines whether the received measured temperature is within the predetermined temperature threshold. If the received measured temperature is within the threshold, control device 130 may continue performing method 300 by returning to step 310. If the received measured temperature is outside the threshold, control device 130 may proceed to step 340.

At step 340, control device 130 takes corrective action in an attempt to bring the measured temperature back within the required range, being the threshold of step 320. The corrective action may involve changing the operation of components of system 100. For example, control device 130 may adjust the operation of fluid control mechanism 127 to cause more or less heat to be dissipated from the first cooling fluid to the second cooling fluid, or from the second cooling fluid to a third cooling fluid. In some embodiments, control device 130 may do this by altering the operation of the fluid control mechanism 127 to increase or decrease the flow rate of the second cooling fluid or the third cooling fluid, respectively. Where fluid control mechanism 127 comprises a pump, changing the flow rate of the second cooling fluid or the third cooling fluid may be achieved by altering the operation of the pump. Where fluid control mechanism 127 comprises a valve, changing the flow rate of the second cooling fluid or the third cooling fluid may be achieved by at least partially opening or closing the valve. Where fluid control mechanism 127 comprises a fan, changing the flow rate of the second cooling fluid or the third cooling fluid may be achieved by altering the operation of the fan. Control device 130 may then return to step 310, to receive a further temperature measurement. The method 300 may be performed iteratively to adjust the measured temperature, and to maintain the measured temperature within the desired temperature range.

Figure 4 shows a schematic diagram of an alternative cooling system 400 for cooling computing devices 103. Cooling system 400 shares many components with cooling system 100, which are shown in a simplified form compared to Figure 1. While certain components of system 100 are not shown in Figure 4 for simplicity, system 400 may comprise any components described above with reference to Figures 1 or 2. For example, system 400 may comprise a control device 130.

As well as the tanks 101, first fluid reservoir 120 and second fluid reservoir 140 as described above with reference to Figures 1 and 2, system 400 comprises a number of redundant components. Specifically, system 400 comprises a first set of cooling components 410 comprising pump system 110A, heat dissipation system 160A, heat exchanger 115A, and heat dissipater 125 A, and a second set of cooling components 420 comprising pump system 110B, heat dissipation system 160B, heat exchanger 115B, and heat dissipater 125B. While two sets of cooling components are illustrated, system 400 may have any number of redundant components, such as three, four or more sets of redundant components.

According to some embodiments, the first set of cooling components 410 may operate in a different manner than the second set of cooling components 420. For example, the first set of cooling components 410 may have a second cooling fluid that is water, which is passed through heat exchanger 115A to absorb heat from the first cooling fluid, and then passed through the heat dissipater 125A to cause the heat from the water to be dissipated. The dissipating may be into a third cooling fluid, which may be air. The second set of cooling components 420 may have a second cooling fluid that is air, which is passed only through heat dissipater 125, to cause the heat to be dissipated into the air and then cause the air to be released into the atmosphere. In such an embodiments, the second set of cooling components 420 may not include a heat exchanger 115B. Such an embodiment is described in further detail below with reference to Figure 6.

Further, while Figure 4 shows certain components as having redundancy and certain components as being single shared components, a skilled person will recognise that any component shown may be installed with or without a redundancy. For example, while only a first fluid reservoir 120 and second fluid reservoir 140 are shown, according to some embodiments, system 400 may include multiple first fluid reservoirs 120 and/or second fluid reservoirs 140 to act as redundant reservoirs.

In Figure 4 two sets of conduits 150 and 156 are shown. Two sets of pipework including conduits 150 and 156 may be used to provide a completely “N+N” redundant system with separate pipework for each set of cooling components 410/420. For example, this may include separate inlet and outlet conduits 150 and 156 connecting to each tank 101, rather than a shared supply and/or return header being used. Implementing a discrete N+N solution for redundancy allows for a concurrently maintainable system with reduced complexity and cost. In some alternative embodiments, system 400 may comprise only one set of each of these conduits connecting between the components 410/420 and the tanks 101.

According to some embodiments, pump systems 110A and HOB may be operatable to cause the first cooling fluid to move through either or both of the first set of cooling components 410 and/or the second set of cooling components 420. In some embodiments, each of pump systems 110A and HOB may be controllable by control device 130 to operate at predetermined percentages or proportions of their operational capacity. According to some embodiments, each of pump systems 110A and 110B may be controllable by control device 130 to operate at a different percentage or proportion of its operational capacity.

Each pump system 110A and 110B alone may have the operational capacity required to move the first cooling fluid through a respective set of cooling components at a desired flow rate. The power required to do so may be referred to as the maximum required operating capacity of the individual pump system. According to some embodiments, each pump system 110A and HOB may be controllable by control device 130 to operate at predetermined percentages or proportions of their maximum required operating capacity. In other words, each pump system 110A and HOB may be controllable by control device 130 to operate at less than its maximum required operating capacity. For example, each pump system 110A and 110B may be controlled to operate at 0%, 50%, and 100% of its maximum required operating capacity, or at any level between 0% and 100%, in some embodiments.

Where pump system 110A is operating at 100% of its maximum required operating capacity and pump system 110B is operating at 0% of its maximum required operating capacity, the first cooling fluid may be caused to move through the first set of cooling components 410. Where pump system 110A is operating at 0% of its maximum required operating capacity and pump system 110B is operating at 100% of its maximum required operating capacity, the first cooling fluid may be caused to move through the second set of cooling components 420. Where both pump systems 110A and 110B are operating (at more than 0% of their maximum required operating capacity), the first cooling fluid may be caused to move through both the first set of cooling components 410 and the second set of cooling components 420. For example, where pump system 110A is operating at 50% of its maximum required operating capacity and pump system 110B is operating at 50% of its maximum required operating capacity, the first cooling fluid may be caused to move through both the first set of cooling components 410 and the second set of cooling components 420, and may flow through each set of cooling components at a substantially equal flow rate. In some embodiments, each of the first set of cooling components 410 and the second set of cooling components 420 remains fluidly connected to tanks 101 even when the respective pump system is operating at 0% of its maximum required operating capacity, and so these components may still be filled with the first cooling fluid even when non-operational. However, the operation of the pump systems 110A and 110B in the manner described may cause the first cooling fluid to remain substantially stationary in the set of cooling components that is non-operational (where the respective pump is not operating or is operating at 0% of its maximum required operating capacity), and to only move through the set of cooling components that are operational (where the respective pump is operating at more than 0% of its maximum required operating capacity). According to some embodiments, during normal operation only the first set of cooling components 410 may be in operation, with pump system 110A running at 100% of its maximum required operating capacity. The second set of cooling components 420 may be non-operational, with pump system HOB running at 0% of its maximum required operating capacity. The first cooling fluid may therefore move through pump system 110A and heat exchanger 115 A. Any of the first cooling fluid located in pump system 110B or heat exchanger 115B may be substantially stationary. If a fault is detected in the first set of cooling components 410, pump system 110A may be turned off and pump system HOB may be turned on. This will cause the first set of cooling components 410 to become non-operational, with pump system 110A running at 0% of its maximum required operating capacity, and the second set of cooling components 420 to become operational, and with pump system 110B running at 100% of its maximum required operating capacity. The first cooling fluid may therefore start to move through pump system 110B and heat exchanger 115B, and any of the first cooling fluid located in pump system 110A or heat exchanger 115A may cease moving and become substantially stationary.

In an alternative embodiment, during normal operation both sets of cooling components may be in operation, with pump system 110A running at 50% of its maximum required operating capacity and pump system 110B running at 50% of its maximum required operating capacity. The first cooling fluid may therefore move through both pump system 110A and heat exchanger 115A as well as pump system 110B and heat exchanger 115B. If a fault is detected in either set of cooling components, the respective pump system may be turned off and the other pump system may be turned on, or have its operation capacity increased to 100% of its maximum required operating capacity. This will cause one set of cooling components to become non-operational, with the respective pump system running at 0% of its maximum required operating capacity, and the other set of cooling components to become operational, with the respective pump system running at 100% of its maximum required operating capacity. The first cooling fluid may therefore cease moving through the non-operational cooling components, and move only through the operational cooling components.

This alternative embodiment may be preferable in some cases, as running both pump systems at 50% of their maximum required operating capacity may increase longevity of the pump systems, as each pump system is placed under less stress. The response time in case of a fault may also be quicker, as it is quicker to ramp up a pump system from 50% power to 100% power than to ramp up from 0% to 100%. Furthermore, this embodiment may have increased efficiency, as having two heat exchangers in operation will result in a higher radiance of heat for the energy used to run the pumps, and will therefore increases the power usage effectiveness (PUE) of the system.

While the above examples have been described with reference to the pump systems 110 operating at 0%, 50% or 100% of their maximum required operating capacity, it should be understood that similar results can be achieved by the pump systems 110 operating at other levels. For example, a pump system 110 may operate at between 0% and 30% of its maximum required operating capacity rather than exactly 0%. A pump system 110 may operate at between 30% and 70% of its maximum required operating capacity rather than exactly 50%. A pump system 110 may operate at between 70% and 100% of its maximum required operating capacity rather than exactly 100%.

There are a number of ways a fault may be detected in the sets of cooling components. In some embodiments, pressure sensors may be placed in one or more locations in system 400, and a drop in pressure below a predetermined threshold may indicate a fault. However, where the first cooling fluid being used in system 400 is an oil, this may be difficult to implement, as pressure sensors often comprise rubber bellows which will deteriorate in oil. Furthermore, pressure sensors require penetrations in the pipework and become a serviceable item. As the pressure sensors are integrated with the pipework, the system would need to be shut down and isolated for the servicing to take place.

In some embodiments, flow meters may be used to detect faults in the system. However, conventional flow meters may not be able to be used as the materials they are made of may not be compatible with the cooling fluid or may disintegrate in the cooling fluid. Instead, ultrasonic flow meters may be used. One or more ultrasonic flow meters may be positioned externally to the pipework to measure flow and detect faults in the system by identifying when the measured flow is outside of normal operating ranges. The use of such flow meters positioned outside the pipework reduces the number of flanges and provides for a more resilient system.

In some embodiments, faults may additionally or alternatively be detected by measuring aspects of the performance of the pump systems 110A and HOB. Specifically, faults may detected by monitoring pump sensors 116A and 116B of the pump systems 110A and HOB. According to some embodiments, these pump systems may comprise a variable speed drive (VSD) speed controller for controlling the speed of the pumps. A VSD may be configured to control the speed and torque of a pump motor, and may comprise a number of sensors to detect the performance of the pump motor. For example, a VSD may include a motor torque sensor for sensing the torque of the pump motor. In such embodiments, pump sensors 116A and 116B may comprise one or more sensors of the VSD, such as the VSD motor torque sensor, for example. According to some embodiments, the pump sensors 116A and 116B may comprise motor torque sensors independent of a VSD. The torque sensors may be used to detect a fault within the sets of cooling components 410 and/or 420. Specifically, a fault may be determined to have occurred when a pump motor is detected as operating at less than a predetermined percentage of its ordinary running torque. For example, a fault may be determined to have occurred where a pump motor is detected as operating at less than 70% of its ordinary running torque. In some embodiments, a fault may be determined to have occurred where a pump motor is detected as operating at less than 85% of its ordinary running torque. In some embodiments, a fault may be determined to have occurred where a pump motor is detected as operating at less than 90% of its ordinary running torque.

By using motor torque to determine if the pump is operating in an unusual manner, system faults can be identified in a non-invasive manner that does not require further penetrations to be made in the pipework.

In some further embodiments, faults may be detected by measuring aspects of the performance of other components or elements of system 100. For example, a fault may be determined to have occurred when a temperature measured at a location within system 100 is outside of a predetermined temperature range associated with that location. In some embodiments, a fault may be determined to have occurred when a status of a component of system 100 is determined to be “off’, or where the component is otherwise determined to be non-operational. In some embodiments, a fault may be determined to have occurred when a valve position or configuration is determined to be outside the range required of the valve for operation. For example, a fault may be determined to have occurred when the valve position is determined to be open when the valve should be closed for proper operation. In some embodiments, a fault may be determined to have occurred when the valve position is determined to be closed when the valve should be open for proper operation. Figure 5 shows a flowchart illustrating a method 500 that may be carried out by control device 130 to operate a redundant cooling system such as system 400 and to switch between the redundant sets of cooling components 410/420.

At step 510, control device 130 receives a pump sensor measurement value relating to an operation of a pump motor within system 400. For example, the value may be measured within pump system 110A and/or pump system HOB. The measurement may be received from a sensor located in or near the respective pump system, which may be part of a VSD controlling the pump system in some embodiments. According to some embodiments, the sensor may be a motor torque sensor, and the value may be a measurement relating to a motor torque measured at a pump motor within system 400.

At step 520, control device 130 compares the received measurement value to a predetermined threshold. The threshold may be retrieved from a memory storage location accessible to control device 130, in some embodiments. Where the measurement value is a motor torque measurement, the predetermined threshold may be a percentage of a normal operating torque of the respective pump motor in some embodiments. In some embodiments, the torque threshold may be a range between 0% and 70% of ordinary operating torque. In some embodiments, the torque threshold may be a range between 0% and 85% of ordinary operating torque. In some embodiments, the torque threshold may be a range between 0% and 90% of ordinary operating torque.

At step 530, control device 130 determines whether the received measurement value is within the predetermined threshold. If the received measurement value is within the threshold, control device 130 may continue performing method 500 by returning to step 510. If the received measurement value is outside the threshold, control device 130 may proceed to step 540.

At step 540, control device 130 causes the pump motor corresponding to the received measurement value to cease operating, such that the operational power of the pump motor is brought down to 0% of its maximum required operating capacity.

At step 550, control device 130 causes the redundant pump motor to begin operating at full power, such that the operational power of the redundant pump motor is brought up to 100% of its maximum required operating capacity. While the phrase “redundant pump motor” has been used here, as described above, in some embodiments this pump motor might already be operating at 50% of its maximum required operating capacity, and so may be redundant only in so much as it can take on 100% of the operation when required due to a detected fault.

Method 500 may be performed in parallel for each pump system 110A/110B which is in operation.

Figure 6 shows a schematic diagram of a specific embodiment of cooling system 400 according to some embodiments.

Figure 6 shows three tanks 101 connected to a first set of cooling components including conduits 150A, 152A and 156A. Conduit 150A allows heated first cooling fluid to escape from tanks 101, and fluidly connects tanks 101 with fluid pump 112A. Conduit 152A carries the first cooling fluid from fluid pump 112A to heat dissipater 125A. Conduit 156A returns cooled first cooling fluid back to tanks 101.

The tanks are also connected to a second set of cooling components including conduits 150A, 152B and 156B.. Conduit 150B allows heated first cooling fluid to escape from tank 101, and fluidly connects tank 101 with fluid pump 112B. Conduit 152B carries the first cooling fluid from fluid pump 112B to heat exchanger 115B. Conduit 156B returns cooled first cooling fluid back to tanks 101..

Heat exchanger 115B facilitates the dissipation of heat from the first cooling fluid delivered by conduit 152B into a second cooling fluid, being water in the illustrated embodiment. The cooled first cooling liquid is then returned to tanks 101 via conduit 156B. Conduit 158B delivers the water to heat exchanger 115B, and conduit 159B returns the heated water that has absorbed the heat from the first cooling liquid. A fluid pump 610 works to pump the water through conduits 158B and 159B, between heat exchanger 11B and heat dissipater 125B.

Heat dissipater 125 A is an adiabatic cooler in the illustrated embodiment. Heat dissipater 125 A facilitates the dissipation of heat from the first cooling fluid delivered by conduit 152A into a second cooling fluid, being air in the illustrated embodiment. The air is then released into the environment. Heat dissipater 125B is also an adiabatic cooler in the illustrated embodiment. Heat dissipater 125B facilitates the dissipation of heat from the second cooling fluid, being water, delivered by conduit 159B. This may be by dissipating the heat from the water into a third cooling fluid such as air, for example. Cooled water from heat dissipater 125B is then returned via conduit 158B .

It will be appreciated by persons skilled in the art that numerous variations and/or modifications may be made to the above-described embodiments, without departing from the broad general scope of the present disclosure. The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive.