Power Distribution Units (PDUs) for optimizing the distribution of power to multi-PSU ICT equipment such as servers and network switches deployed in data center racks
The power distribution system addresses inefficiencies and cybersecurity issues in data center PDUs by dynamically balancing power phases and providing redundant power sources with rapid switching, optimizing power use and reducing heat loss.
Patent Information
- Application Number
- JP2025530276
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2022-11-28
- Filing Date
- 2023-11-28
- Publication Date
- 2025-11-28
AI Technical Summary
Existing power distribution units (PDUs) in data centers suffer from suboptimal power consumption and introduce cybersecurity flaws due to current imbalances, leading to excess capacity, heat loss, and inefficiencies in power distribution to multi-PSU servers.
A power distribution system with phase switches and sensors that dynamically balance power consumption across phases, incorporate fault prediction, and provide redundant power sources with rapid switching to minimize inefficiencies and enhance cybersecurity.
The system optimizes power consumption by equalizing power use across phases, reduces heat loss, and ensures reliable power delivery with minimal redundancy risks, enhancing overall efficiency and security.
Smart Images

Figure 2025538593000001_ABST
Abstract
Description
[Technical Field]
[0001] The present invention relates to a power distribution unit (PDU) module, a power bar comprising one or more PDU modules, and a power distribution system comprising a plurality of power bars. The present invention further relates to the use of the above-mentioned elements. [Background technology]
[0002] Centralized data centers for servers, network switches, and other information and communications technology (ICT) equipment have been in use for many years. A typical centralized data center contains many racks of ICT equipment that require power. A power distribution unit (PDU) is a device typically used to distribute power from a power input to the ICT equipment housed in the racks.
[0003] However, existing PDU implementations provide suboptimal power consumption and typically introduce cybersecurity flaws.
[0004] Current racks contain multi-PSU servers (PSU stands for power supply unit). The distribution of power to these multi-PSU servers is also suboptimal, resulting in suboptimal power consumption for the rack. Power distribution systems with three-phase inlets and hardwired single-phase outlets inherently create current imbalances in the computer room's power distribution grid. Phase-to-phase imbalances result in excess capacity for both power and ICT. Furthermore, imbalances cause upstream heat loss in the cabling leading to the uninterruptible power supply (UPS). UPSs also generate more heat because they must compensate for the imbalance introduced at the rack level. Furthermore, considering that load imbalances change over time as IT workloads change, improvements are needed to account for this change over time as IT workloads change.
[0005] Therefore, based on the above, there is a need for new technologies that can optimize power consumption and have limited cybersecurity flaws. Summary of the Invention
[0006] As such, the present invention offers several improvements over the prior art.
[0007] In one embodiment of the present invention, a computer-implemented method for distributing three-phase power in a rack system includes a power distribution system for powering electrical equipment configured to be supplied with a three-phase power source, the power distribution system including at least one power bar with at least three outlets, a power sensor for measuring voltage and current at each outlet, and a phase switch for switching the live conductors of the outlets, the method comprising: receiving voltage and current values for each outlet; calculating the power of each outlet based on the received voltage and current values; registering the calculated power of each outlet and the corresponding phase setting; calculating a power load imbalance based on the calculated power and phase configuration of each outlet to obtain a calculated load imbalance; calculating individual power calculations at each outlet via an iterative process, where load imbalance is calculated for hypothetical combinations of phase configurations; selecting the hypothetical combination with the lowest load imbalance as the best case; Comparing the best case load imbalance with the calculated load imbalance; if the best case load imbalance is lower than the calculated load imbalance, sending an instruction to the phase switch to set the phase setting to the best case phase setting; A computer-implemented method is provided, comprising:
[0008] Equalizing power consumption across the three phases maximizes excess power and eliminates excess capacity, while reducing upstream heat losses.
[0009] In one embodiment of the present invention, a power distribution system is provided that includes a processor configured to perform an imbalance correction method.
[0010] In one embodiment of the present invention, a computer program is provided comprising instructions that, when the program is executed by a computer, cause the computer to perform the steps of an imbalance correction method.
[0011] In one embodiment of the present invention, there is provided a further computer-implemented method for distributing power in a rack system, the rack system comprising a power distribution system for powering multi-PSU electrical equipment, the rack system being supplied with at least two power sources, the power distribution system comprising at least two power bars with at least one outlet, a power sensor for measuring voltage and current signals of each outlet, and a decoupling switch for switching live wires to the outlets on or off, the method comprising: receiving voltage and current signals for each outlet; applying power quality measurements to the received signal to determine a trigger for a power outage; if a power outage is determined to have been triggered for an outlet connected to the multi-PSU equipment, determining another outlet connected to the multi-PSU equipment; sending a command to the other outlets to switch their decoupling switches to an on state; Further computer-implemented methods are provided, including:
[0012] In one embodiment of the present invention, there is provided a power distribution system comprising a processor configured to perform the power outage correction method described above.
[0013] In one embodiment of the present invention, a computer program is provided that includes instructions that, when the program is executed by a computer, cause the computer to perform the steps of a method for correcting a power outage.
[0014] In one embodiment of the present invention, a power distribution system is provided that overcomes the shortcomings of the prior art.
[0015] The advantage of this system is that it can switch the outlet module completely or individual outlets by phase. For example, a C13 module may switch all outlets of the module, and a C19 module may switch individual outlets of the module.
[0016] This system also has the advantage that the controller can be configured to perform load imbalance correction.
[0017] Above all, the system has the advantage of providing dynamic load redistribution capabilities at the ICT equipment level.
[0018] In one embodiment of the present invention, there is provided a PDU module comprising: an input coupled to the power source and configured to receive three-phase power from the power source; a plurality of power lines coupled to the input, the plurality of power lines configured to carry three-phase power; a plurality of outlets each coupled to two of a plurality of power lines to receive power from one of three phases of three-phase power, each outlet configured to be coupled to an electrical device to power the electrical device; Preferably, each outlet is provided with a power sensor configured to sense a parameter of the power supplied to said outlet, the sensed parameter being usable to adapt the power delivered to the outlet.
[0019] The PDU module includes a controller coupled to a power sensor.
[0020] In one embodiment, the PDU module further comprises a data connector configured to receive data from an external device, such as another PDU module coupled to the same power source. In an alternative embodiment described below, multiple PDU modules can be plugged into a power bar. In this case, data connectors can be coupled to some or all of the PDU modules. The data connectors can be provided, for example, on a dedicated gateway module plugged into one power bar and configured to receive data from an external device, such as a PDU module in another power bar coupled to a different power source.
[0021] In one embodiment, the PDU module further comprises a controller. In an alternative preferred embodiment described below, multiple PDU modules can be assembled into a power bar. In this case, the controller can be centralized for some or all of the PDU modules. The controller can be provided, for example, in a dedicated gateway module provided in the power bar, e.g., the same gateway module that provides the centralized data connector.
[0022] According to a first refinement of the present invention, the PDU module includes the features of claim 2. The PDU module thus further comprises at least one phase switch configured to selectively couple each outlet to two of the power lines to select the phase delivered to the outlet. Preferably, the three-phase power line comprises a neutral conductor and three live conductors. Preferably, only the live conductors are switched by the phase switches, i.e., the neutral conductor is not switched. This makes it possible, for example, to provide a PDU module that operates in a wye configuration (as opposed to a delta configuration), with the outlet operating voltage being taken between the live conductors and the neutral conductor.
[0023] According to one embodiment of the present invention, the phase switch is further configured to selectively decouple the outlets from the power line. To this end, the phase switch preferably comprises a switching section and a decoupling section. The switching section performs the above-mentioned switching of the power line. The decoupling section breaks the power line. This allows for a "break-before-make" configuration to be implemented, allowing for "live switching".
[0024] According to one embodiment of the present invention, one phase switch is provided for each outlet to select the phase delivered to each outlet. However, alternatives are possible. For example, one phase switch may be provided for a group of outlets to select the phase delivered to the group of outlets, for example, all the outlets on a PDU module. Alternatively, for example, only the switching section may be common to the group of outlets, and each outlet may be provided with, for example, its own decoupling section.
[0025] According to one embodiment of the present invention, a parameter measured by a power sensor of an outlet indicates a load on the outlet, also referred to as a load parameter. Preferably, the load parameter is a current drawn by the outlet. Preferably, the controller is configured to monitor the load parameters of the outlets in the PDU module and is further configured to operate at least one phase switch based on the monitored load parameters. Preferably, the controller determines a total load on each of the phases based on the load parameters, and the controller is further configured to operate at least one phase switch to improve load balancing of the three phases. Improving load balancing of the three phases has many advantages. One advantage is particularly present when the PDU is supplied with power from a power source via an uninterruptible power supply (UPS), in which case a load imbalance causes a lightly loaded phase to continuously charge the UPS and a heavily loaded phase to discharge the UPS. Preferably, the controller maintains a list of outlets, the phases delivered to the outlets, and the load parameters of the outlets, and the controller uses the list to determine which outlets should switch phases, and the controller operates the phase switches accordingly. Preferably, the controller determines the highest-load phase and the lowest-load phase and uses the list to determine which outlets should switch from the highest-load phase to one of the lowest-load phases. Preferably, the list further includes an outlet priority indicating the importance of reliable operation of the outlet. The controller is preferably configured to switch those outlets having a higher priority status to a more stable phase. According to one embodiment of the present invention, the controller is configured to receive, via the data connector, phase load information regarding three-phase loads of at least one external device coupled to the same power source, and is further configured to operate the at least one phase switch further based on the received phase load information. Preferably, the external device coupled to the power source is another PDU module coupled to the same power source. This embodiment is particularly advantageous when dealing with a power bar comprising multiple PDU modules, as will be further described below.
[0026] According to a second refinement of the invention, a PDU module, such as the PDU module of the first refinement of the invention described above, comprises the features of claim 13. In the second refinement, the power sensor is enabled to sample the load parameter at the microsecond level, and the controller is adapted to enable an oscilloscope diagram graphical representation of the parameter in the time domain and / or the frequency domain, the representation preferably including a representation of harmonics of the parameter.
[0027] According to one embodiment of the present invention, the controller is configured to identify anomalies in the time-domain and / or frequency-domain representations of the parameters. This embodiment implements a fault prediction system. The fault prediction system uses reference oscilloscope graph representations of a properly functioning PSU (power supply unit) of an electrical device and compares these references with real-time oscilloscope graph representations of the parameters. If the differences between the representations become significant or rapidly evolving, this can be an indicator of wear of the electrical device or PDU module. The reference oscilloscope graph representation may be, for example, a snapshot taken at the time of commissioning the electrical device or PDU module, or may be obtained, for example, from a library of known electrical devices and PDU modules. The comparison may be performed, for example, by machine learning.
[0028] According to one embodiment of the present invention, the controller is configured to compare the measured parameters with reference data retrieved from a cloud in which the measured parameters of the PDU module, preferably measured parameters from other PDU modules, are stored, so as to enable identification of anomalies in the time domain and / or frequency domain representation of the parameters, for example based on machine learning.
[0029] According to one embodiment of the present invention, the parameter measured by the power sensor is the current delivered to the outlet. According to one embodiment of the present invention, the input of the PDU module further comprises a voltage sensor configured to measure the voltage between the power lines at the microsecond level, and the controller is adapted to enable an oscilloscope diagram graphical representation of the voltage in the time and / or frequency domain, preferably including a representation of the harmonics of the parameter, i.e. mutatis mutandis for the analysis performed based on the above mentioned power sensor.
[0030] According to one aspect of the present invention, a PDU module, such as a PDU module according to the first and / or second improvements of the present invention described above, includes the features of claim 17, i.e., the PDU module is integrated into a power bar. The power bar comprises an elongated frame configured to be mounted in a data center rack in a 0U or 1U configuration. Preferably, the power bar is mounted vertically in a 0U configuration. The frame is also referred to as a "backbone." The power bar comprises at least one PDU module as described above mounted in the frame. The power bar allows the PDU module to be modularly and removably mounted to the frame, for example, allowing for the replacement of a PDU module including C19 outlets with a PDU module including C13 outlets. According to one embodiment of the present invention, a controller and / or data connectors are attached to the power bar, for example, in a removably manner. Preferably, the power bar comprises a data connector and / or controller common to all PDU modules in the power bar. The centralized data connector and centralized controller are preferably provided in a dedicated gateway module. The gateway module is preferably modularly and removably mounted to the frame. Preferably, the gateway module has ports for connecting to environmental sensors, such as temperature and humidity sensors, that provide the controller with information about environmental conditions within the data center. These conditions may affect, for example, the oscilloscope diagram graphical representation of the load parameters, and are therefore preferably taken into account when comparing the graphical representation with a reference.
[0031] According to one embodiment of the present invention, the power bar further comprises at least one network switch module having a plurality of network ports, the at least one network switch module being removably mounted to the frame. Thus, the modularity of the power bar extends beyond its ability to removably receive one or more PDU modules, and preferably extends to the network switch modules it removably receives. According to one embodiment of the present invention, the power bar comprises a plurality of PDU modules. According to one embodiment of the present invention, the frame comprises a power inlet connected to a power source. Modules mounted in the frame, i.e., modules as described above, receive power from the power inlet.
[0032] According to one embodiment of the present invention, the frame includes a power bus, and modules mounted on the frame connect to the power bus to transfer power from the power bus to the modules. Preferably, the modules can be inserted into the frame such that the power contacts of the modules automatically make contact with the power bus.
[0033] According to one embodiment of the present invention, the frame includes a data bus over which the modules can exchange data. Preferably, the data bus communication is Ethernet communication. This allows the modules to communicate in a peer-to-peer manner, as opposed to a master-slave manner. Preferably, the modules can be inserted into a power bar where the data contacts of the modules automatically make contact with the data bus.
[0034] According to a further improvement of the present invention, power consumption is reduced by tightly coupling a power distribution system comprising a first power bar as described above configured to couple a set of electrical equipment to a first three-phase power source and a second power bar as described above configured to couple the same set of electrical equipment to a second three-phase power source, the set of electrical equipment having two physically different power inlets, the first power inlet connected to the first three-phase power source and the second power inlet connected to the second three-phase power source. Thus, the redundant power delivery system is a multiple-feed system, e.g., a dual-feed system, in which a rack is provided with multiple, e.g., two independent power sources, and the electrical equipment has multiple, e.g., two power inlets, each powered by a separate power bar. While this multiple-feed system ensures that the electrical equipment is substantially always powered, it results in excess power consumption because some electrical circuits are implemented in multiples to provide redundancy, simultaneously weakening the load and increasing inefficiency.
[0035] According to one embodiment of the present invention, a system includes a power switching controller configured to switch power delivery to one of a set of electrical devices having outlets in a first power bar and outlets in a second power bar. Preferably, the switching includes splitting power delivery to the electrical device between the outlets in the first power bar and the outlets in the second power bar according to a splitting scheme. Preferably, the splitting scheme includes powering the electrical device entirely by outlets in one of the first power bars (referred to as primary outlets) and disabling outlets in the second power bar. This embodiment ensures that the amount of power consumed by the electrical device is minimized. This naturally leads to redundancy risks, which can be controlled by providing a rapid switchover from a failed outlet to a working outlet on another power bar. Thus, one embodiment includes, upon detecting a failure of a primary outlet, switching the electrical device to be powered entirely by the other outlet of the first and second power bars, referred to as backup outlets.
[0036] In one embodiment, all outlets in the first power bar are primary outlets and all outlets in the second power bar are backup outlets. To ensure very fast switching, the first power bar and the second power bar are directly connected by a data connector, preferably a data cable, to increase the speed at which the power switching controller can switch the outlets powering the electrical devices. This avoids the need to send signals from the first power bar to the second power bar over an external network path. Avoiding the external network path simultaneously enhances cybersecurity.
[0037] According to one embodiment of the present invention, the power switching controller is located in either the first power bar or the second power bar, but not both. This results in a less expensive redundant power delivery system. Preferably, the power switching controller is located in the gateway module of the power bar. For maximum switching speed, the switching controller is implemented in the PDU module containing the primary outlet. This controller monitors power parameters at the microsecond level, continuously checking time and frequency domain measurements, including harmonics, and alerts the PDU module containing the backup outlet in a peer-to-peer manner when an anomaly is detected.
[0038] It is a further object of the present invention to provide a data center rack comprising at least one power bar as described above having at least one PDU module as described above mounted thereon, or at least one redundant power delivery system as described above.
[0039] It is a further object of the present invention to provide a method comprising the use of the PDU module as described above, or the power bar as described above, the power distribution system as described above, or the data center rack as described above.
[0040] The present invention is further explained by the following description and accompanying drawings. [Brief explanation of the drawings]
[0041] [Figure 1] FIG. 1 is a perspective view of a power bar according to one embodiment of the present invention. [Figure 2] 2 shows the power bar of FIG. 1 with the modules removed from the frame; [Figure 3A] 3 shows a frame profile of the frame of the power supply bar shown in FIGS. 1 and 2. FIG. [Figure 3B] 1 shows a cross-section of a frame profile according to one embodiment of the present invention; [Figure 4A] FIG. 2 is a schematic diagram of the wiring of a PDU module according to one embodiment of the present invention. [Figure 4B] 1 is an alternative wiring of a PDU module according to one embodiment of the present invention. [Figure 4C] 10 is a further alternative wiring of a PDU module according to one embodiment of the present invention. [Figure 5A] 1 shows a graphical representation of load parameters in the time domain for a reference healthy PSU (power supply unit). [Figure 5B] 1 shows a graphical representation of load parameters in the time domain for a reference healthy PSU (power supply unit). [Figure 5C] 1 shows a graphical representation of load parameters in the time domain for a reference healthy PSU (power supply unit). [Figure 5D] 1 shows a graphical representation of load parameters in the time domain for a failed PSU (power supply unit). [Figure 5E] 1 shows a graphical representation of load parameters in the time domain for a failed PSU (power supply unit). [Figure 5F] 1 shows a graphical representation of load parameters in the time domain for a failed PSU (power supply unit). [Figure 6] 1 is a schematic diagram of a power distribution system according to one embodiment of the present invention; [Figure 7] 1 shows a flowchart of a load imbalance algorithm that may be executed by the controller 17 of the power distribution system 60. [Figure 8] FIG. 1 is an exploded view of the frame of the power bar. [Figure 9] 1 shows a perspective view of a PDU module. [Figure 10] A dual PSU server is shown with the decoupling switch in four different states. [Figure 11] 1 illustrates the power consumption of a dual PSU server over a 48 hour period as controlled by an algorithm according to one embodiment of the present invention. [Figure 12] The graph shows the efficiency of the PSU on the y-axis for each percentage of load on the x-axis. [Figure 13]10 illustrates the difference in heat loss between a conventional system without imbalance correction and a system according to an embodiment of the present invention with imbalance correction. [Figure 14] 10 illustrates the difference in heat loss between a conventional system without imbalance correction and a system according to an embodiment of the present invention with imbalance correction. [Figure 15] 10 illustrates the difference in heat loss between a conventional system without imbalance correction and a system according to an embodiment of the present invention with imbalance correction. [Figure 16] 1 illustrates a computer-implemented method for fault checking according to one embodiment of the present invention. [Figure 17] 1 illustrates a computer-implemented method for fault checking according to one embodiment of the present invention. [Figure 18] 1 illustrates a computer-implemented method for fault checking according to one embodiment of the present invention. [Figure 19] 1 illustrates a computer-implemented method for fault checking according to one embodiment of the present invention. [Figure 20] 1 shows the detection of the expected zero crossing time shift of the voltage curve over time. DETAILED DESCRIPTION OF THE INVENTION
[0042] BRIEF DESCRIPTION OF THE DRAWINGS AND MODES FOR CARRYING OUT THE INVENTION
[0043] The present invention will be described with respect to particular embodiments and with reference to certain drawings but the invention is not limited thereto but only by the claims. The drawings described are schematic only and are non-limiting. In the drawings, the size of some of the elements may be exaggerated and not drawn to scale for illustrative purposes. The dimensions and relative dimensions do not necessarily correspond to reductions to actual implementations of the invention.
[0044] Furthermore, the terms first, second, third, etc. in the specification and claims are used to distinguish between like elements and do not necessarily describe a sequential or chronological order. These terms are interchangeable under appropriate circumstances, and embodiments of the invention may operate in sequences other than those described or illustrated herein.
[0045] Furthermore, terms such as top, bottom, upper, lower, etc. in the specification and claims are used for descriptive purposes and not necessarily to describe relative positions. Terms so used are interchangeable under appropriate circumstances, and embodiments of the invention described herein may operate in orientations other than those described or illustrated herein.
[0046] Moreover, various embodiments, while referred to as "preferred," should not be construed as limiting the scope of the invention, but as exemplary ways in which the invention may be practiced.
[0047] The term "comprising" as used in the claims should not be construed as being limited to the subsequently listed elements or steps, and does not exclude other elements or steps. It should be interpreted as specifying the presence of the stated features, integers, steps, or components as referenced, but does not exclude the presence or addition of one or more other features, integers, steps, or components, or groups thereof. Thus, the scope of the expression "a device comprising A and B" should not be limited to a device consisting only of components A and B; for the present invention, the only listed components of the device are A and B, and the claims should be interpreted to include equivalents of those components.
[0048] Furthermore, in the context of this specification, the terms "imbalance" and "unbalance" are used as equivalent terms with the same meaning.
[0049] 1-3 show a power bar 2. The power bar comprises an elongated frame 3 configured to be mounted in a data center rack in a 0U configuration (0U corresponds to not being mounted in a slot reserved for a 19-inch rack-mountable device, and therefore not using a unit U of the 19-inch rack). The frame 3 is also referred to as a "backbone." The power bar comprises a plurality of PDU modules 1 mounted in the frame 3.
[0050] As specifically shown in FIG. 4A , each PDU module 1 includes an input 5 configured to be coupled to a power source to receive three-phase power from the power source. A plurality of power lines (51, 52, 53, 54) are coupled to the input 5, with the power lines (51, 52, 53, 54) configured to carry three-phase power. The power lines are three live lines (51, 52, 53) and one neutral line 54. A plurality of outlets 6 are coupled to two of the plurality of power lines to receive power from one of three phases of the three-phase power. In the embodiment of FIG. 4A , the outlets 6 are coupled to one of the three live lines (51, 52, 53) and the neutral line 54. Each outlet 6 is configured to be coupled to an electrical device, such as a server as shown in FIG. 6 , to power the electrical device. Each outlet 6 is provided with a power sensor 30 configured to sense a parameter of the power supplied to at least one outlet 6. The sensed parameter can be used to adapt the power delivered to the outlet 6. The power sensor 30 can include a current sensor 8, a voltage sensor 9, or both. The PDU module 1 also includes a microcontroller 17 coupled to the power sensors 8, 9. The microcontroller 17 is further connected to a data port 18 configured to connect to a second controller 22 via a data connection. Notably, the controller 22 is centralized for all of the PDU modules 1 and is provided within a dedicated gateway module 12, which may be a separate module mounted to the frame 3 as shown in FIGS. 1 and 2. As shown in FIG. 2, the power bar is modular, allowing the PDU modules 1 to be modularly and removably mounted to the frame 3, for example, allowing a PDU module 1 including C19 outlets to be replaced with a PDU module including C13 outlets. Furthermore, the gateway module 12 may also be modularly and removably mounted to the frame. In an alternative embodiment, the gateway module is not modular and is assembled on the frame. In yet another alternative embodiment, the processor 22 may be provided in the PDU module or the frame.The frame 3 of the embodiment of FIGS. 1 and 2 includes a power inlet 11 connected to or configured to be connected to a power source 11. The modules 1, 12 mounted on the frame 3, i.e., modules as described above, receive power from the power inlet 11 when placed in the frame 3 and removably attached. The frame 3 includes power. Alternatively, as shown in FIG. 8, the frame 3 includes several power contact printed circuit boards 28 (PCBs) that are connected to a power source and include pin contact receptacles 23 for receiving some of the pin contacts 25 of the PDU modules 1. The pin contacts 25 are shown in FIG. 9. The power contact PCBs 28 are connected to each other to form a power bus 33. The frame 3 further includes several data PCBs 24 that include data connector receptacles 27 configured to connect with mating data connectors 26 that are part of the modules 1. The mating data connectors 26 are shown in FIG. 9. When the PDU modules 1 are properly installed, the pin contacts 25 of the PDU modules 1 connect to the pin contact receptacles 23, and the data connectors 26 of the PDU modules 1 connect to the data connector receptacles 27 on the PCB boards.
[0051] The modules 1 mounted on the frame 3 connect to the power bus 33 to transmit power from the power bus 33 to the modules. The power bus 33 may include a series of printed circuit boards 28 configured to interface with the PDU modules 1 when the PDU modules 1 are removably mounted on the frame 3. As particularly shown in FIG. 2 , the modules 1 may be mounted on the frame 3 such that the power contacts 25 of the modules 1 automatically contact the power bus 33.
[0052] 3A shows a frame section 32 that is part of the frame 3 of the power bar 2. The frame section 32 includes elongated slots 14 positioned to attach the frame 3 to a data center rack. The provision of slots 14 reduces the risk of short circuits, as opposed to elements that penetrate the frame, such as screws.
[0053] An alternative structure for a frame of a power bar according to one embodiment of the present invention is shown in FIG. 8 . The frame 3 of FIG. 8 according to one embodiment of the present invention includes a frame profile 32 having a particular cross-section. The cross-section of the frame profile 32 is shown in more detail in FIG. 3B . The frame profile 32 has a base wall 91 from which two outer side walls 92, 93 upright to form a generally U-shape. Each outer side wall 92, 93 has two upstanding walls 94, 95 on the inside of the U-shape that form U-shaped grooves 102, 103 in each outer side wall. The frame profile 32 also includes a separation wall 96. The separation wall 96 includes grooves 106, 107 on each side that face the grooves in the outer walls. The groove 106 in the separation wall 96 faces the groove 102 in the outer wall 92. These two grooves 106, 102 are positioned and dimensioned to receive a power contact PCB 28. Groove 107 in separation wall 96 faces groove 103 in outer wall 93. These two grooves 107 and 103 are positioned and dimensioned to receive data PCB 24. Separation wall 96 separates power compartment 104 from data compartment 105. This prevents data signals from being interfered with by power lines, power bars, or power connections.
[0054] FIG. 4A is a wiring schematic diagram of one embodiment of a PDU module 1 within the power bar 2 shown in FIGS. 1-3. The PDU module 1 includes a plurality of phase switches 29 configured to selectively couple each outlet 6 to two of the power lines (51, 52, 53, 54) to select the phase delivered to said outlet 6. In particular, a three-phase power line includes a neutral conductor 54 and three live conductors 51, 52, 53, with only the live conductors being switched by the phase switches 29; i.e., the neutral conductor 54 is not switched as shown in the embodiment of FIG. 4A. The phase switches 29 are further configured to selectively decouple the outlets from the power line. To that end, the phase switches 29 include a switching section 15 and a decoupling section 16. The switching section of the phase switches 29 is also referred to as a three-phase switch 15 in the context of this application. The decoupling section of the phase switches 29 is also referred to as a decoupling switch 16 in the context of this application. The switching unit 15 switches between the above-mentioned live conductors 51, 52, and 53. The decoupling unit 16 breaks the live conductors. This allows for a "break-before-make" configuration and allows for "live switching." In the embodiment of FIG. 4A, the decoupling switch 16 is located between the output contact of the three-phase switch 15 and the outlet 6. In alternative embodiments, the location of the decoupling unit 16 relative to the switching unit 15 may be different. In one embodiment, the decoupling unit 16 may be located before the switching unit 15. For the first three outlets 6 starting from the left in FIG. 4A, one phase switch 29, each including a switching unit 16 and a decoupling unit 15, is provided for each outlet 6 to select the live conductor, if any, delivered to that outlet 6. For the last three outlets 6 starting from the right in FIG. 4A, one phase switch 29 is provided for each group of outlets 6 to select the phase delivered to that group of outlets 6. However, each of the three last outlets 6 from the right is provided with a separate decoupling section 16, so that the phase switch in this configuration comprises one switching section 15 and three decoupling sections 15, i.e. one switching section 15 for one group of outlets and a number of decoupling sections 15 corresponding to the amount of outlets in one group.As shown in FIG. 4A, the power sensor 30 includes a voltage sensor 9 and a current sensor 8 for each of the first three outlets on the left. For the last three outlets from the right of FIG. 4A, the power sensor 30 includes a current sensor 8 for each of the outlets 6 and a voltage sensor 9 common to all three of the outlets 6. The controller 17 is configured to monitor load parameters of the outlets 6 within the PDU module 1, i.e., the voltage from the voltage sensor 9 and the current from the current sensor 8, and is further configured to operate at least one phase switch 29 based on the monitored load parameters. The controller 17 can also determine the total load for each phase based on the load parameters and is further configured to operate at least one phase switch to improve three-phase load balancing, i.e., to make the loads on the three phases as equal as possible. FIG. 4B shows an alternative wiring of the PDU module in which each outlet has a three-phase switch 15 and a decoupling switch 16. In FIG. 4B, each outlet 6 also has a voltage sensor 9 and a current sensor 8. Figure 4C shows yet another wiring of a PDU module where there is only one three-phase switch 15 and one decoupling switch 16 for each outlet. In Figure 4C, there is one voltage sensor 9 for every outlet and one current sensor 8 for each outlet 6.
[0055] 5A, 5B, 5C, 5D, 5E, and 5F show graphical representations of load parameters in the time domain for a reference, healthy PSU (power supply unit) and a faulty PSU, respectively. The microcontroller 17 is configured to detect differences between the graphical representations and determine malfunctions from the detected differences. FIG. 5A shows healthy current and voltage curves over time. FIG. 5D shows a faulty current curve in the same time domain. FIGS. 5B and 5C show current harmonics for a healthy PSU. FIG. 5B shows current amplitude. FIG. 5C shows current phase. A view of the same graphical representation for a faulty PSU is shown in FIG. 5E for current amplitude and in FIG. 5F for current phase. The controller 17 is configured to detect differences between such graphical representations.
[0056] FIG. 6 illustrates a power distribution system 60 according to one embodiment of the present invention for powering a multi-PSU (PSU stands for power supply unit) device 61 in a rack system. The power distribution system 60 includes two power bars 62, 63, referred to as PDU A and PDU B, respectively, each having slots for receiving modules. In one embodiment, the power bars 62, 63 are the power bars of FIGS. 1 and 2. Each power bar includes a power feed 64, 65. The power feeds 64, 65 provide power to the outlets. A first function of the embodiment of FIG. 6 is to operate as a redundant power distribution system. The redundant power distribution system includes a first power bar 62 (PDU A) configured to couple a set of electrical equipment 61 to a first three-phase power source 64 via a first PSU 66 of the electrical equipment, and a second power bar 63 (PDU B) configured to couple the same set of electrical equipment 61 to a second three-phase power source via a second PSU 67. If one of the two power feeds 64, 65 fails, the other can take over.
[0057] Figures 5A, 5B, 5C, 5E, 5D, 5E, and 5F show oscilloscope diagrams of the change in load parameters of the PSUs of electrical equipment, which are load parameters at the corresponding outputs 6 of the PDU modules 1. A first PSU 66 is connected to a first three-phase power source 64, and a second PSU 67 is connected to a second, different three-phase power source 65. Thus, the power distribution system is a dual-power system, with two independent power sources 64, 65 provided for the rack, and the electrical equipment 61 has two PSUs 66, 67, each connected to a separate power bar. This dual-power system ensures that the electrical equipment is substantially always powered. Loss of power from one power source can be backed up by another power source connected to the same electrical equipment. The power distribution system 60 includes a gateway controller 22 in one of the power bars. In the embodiment of Figure 6, the controller is provided in power bar PDU A. The controller 22 is configured to switch the power supply to one electrical device 61 of a set of electrical devices using an outlet 68 provided on PDU A and an outlet 69 provided on PDU B. The switching involves dividing the power delivery to the electrical device between outlets provided on a first power bar and outlets provided on a second power bar according to a division scheme that includes powering the electrical device entirely through outlets provided on PDU A, referred to as primary outlets. This embodiment ensures that the amount of power wasted by standby outlets is minimized. This can naturally lead to redundancy risks, which the present invention controls by quickly switching a failed or failing outlet to a working outlet on power bar PDU B. Some embodiments of the present invention include, upon detecting a failure of a primary outlet, switching the electrical device entirely to powered by outlets on PDU B, which can be seen as a kind of backup outlet. In one embodiment, all outlets on PDU A are primary outlets and all outlets on PDU B are backup outlets.To achieve very fast switching, PDU A and PDU B are directly connected to each other by a data connection 20, which in the embodiment of FIG. 6 is implemented as a data cable. The data connection 20 between the two power bars 62, 63 increases the speed at which the controllers 17, 22 can switch between outlets powering electrical equipment 31. The data connection 20 avoids the need to send signals from the first power bar to the second power bar over an external network path. Avoiding an external network path also enhances cybersecurity. As mentioned above, the gateway controller 22 is located in the gateway module 12 on PDU A only; it is not located on both PDU A and PDU B. Therefore, in the embodiment of FIG. 6, the controller on PDU B is omitted. This is illustrated in FIG. 6 by the controller-less gateway module. This results in a less expensive redundant power distribution system.
[0058] In one embodiment of the present invention, power distribution system 60 is configured to minimize load imbalance between the live lines of a three-phase power feed. In a three-phase system, current imbalance is, for example, the maximum deviation of any phase current from the average divided by the average current. Because the electrical devices connected to power distribution system 60 are single-phase loads, imbalances can occur due to the uneven distribution of single-phase loads across the three phases. Figure 7 shows a flowchart of a load imbalance algorithm that may be executed by controller 22 of power distribution system 60.
[0059] In a first step, for each outlet, the controller calculates the power based on measurements received from voltage sensors 9 and current sensors 8 in the power distribution system 60, and the controller 22 registers the corresponding phase setting for each outlet.
[0060] In the second step, the power load imbalance is calculated based on the calculated power consumption and the information of the phase settings for each outlet, resulting in the calculated load imbalance. The power load imbalance of a three-phase system can be calculated as the maximum deviation of any phase power from the average divided by the average power.
[0061] In the third step, individual power calculations at each outlet are performed via an iterative process, and load imbalances are calculated for hypothetical combinations of phase configurations. For this, an exhaustive list of all possible combinations of phase configurations is obtained by cycling through phase configurations 1, 2, and 3, which correspond to live conductors 51, 52, and 53 in Figure 4A, for each outlet individually.
[0062] In the fourth step, the combination with the lowest calculated load imbalance is selected and labeled as the best case.
[0063] In a fifth step, the best case load imbalance is compared to the calculated load imbalance of the second step, and if the best case load imbalance is lower than the calculated load imbalance ("yes" situation), the controller sends a command to the phase switch 29 to set the phase setting to the best case phase setting. However, if the best case load imbalance is not lower than the calculated load imbalance ("no" situation), no change is made.
[0064] By implementing this method in the power distribution system, load imbalances are minimized and power is used most efficiently, resulting in less energy being consumed by racks running on this power distribution system 60.
[0065] This method of imbalance correction may be performed by a power distribution system having one power bar and one power feed to the power bar, or by a power distribution system having two power bars and two power feeds to the two power bars.
[0066] These power distribution systems can continuously monitor power consumption per outlet. In one embodiment of the present invention, the power imbalance correction described above may be performed once a day at a set time.
[0067] In a power distribution system having two power bars for powering an electrical device with two PSUs, imbalance correction is performed for the outlets of both power bars, but the controller may be configured to perform imbalance correction for each power bar at a different time. This ensures that the electrical device is never powered down because only one of the two PSUs is connected to a power source at any one time. In an alternative embodiment of the invention, the controller may be configured to perform imbalance correction for all outlets of the two bars together and to switch, at different times, the two outlets connected to the two PSUs of an electrical device, such as a server. This, like the previous embodiment, ensures that the electrical device is never powered down because only one of the two PSUs is connected to a power source at any one time.
[0068] Permanently equalizing power consumption across the three phases maximizes excess power and eliminates excess capacity, while reducing upstream heat losses.
[0069] 13, 14, and 15 show the difference in heat loss between a conventional system without imbalance correction and a system according to an embodiment of the present invention with imbalance correction. The reduction in heat loss (HLr) can be defined as the difference between the heat loss (HLt) caused by load imbalance in a hardwired conventional power distribution system and the heat loss (HLi) caused by load imbalance in a power distribution system according to an embodiment of the present invention with imbalance correction. HLr = HLt - HLi
[0070] Measurements show that by using a power distribution system according to an embodiment of the present invention implementing an imbalance correction method according to an embodiment of the present invention, the load imbalance can be improved from over 150% to 50%. This reduction in imbalance corresponds to a reduction in upstream heat loss from approximately 10% to approximately 1%, as shown in Figure 15.
[0071] Figure 2 illustrates the removable installation of a PDU module 1 to a frame 3 according to one embodiment of the present invention. When assembled, the frame includes slots for receiving the modules. The installed PDU module 1 is positioned within the slots such that contact pins 25, shown in Figure 9, mate with contact pin receivers 23 and data connector 26 mates with data connector receiver 27.
[0072] Contact pins 25 connected to contact pin receptacles 23 transmit electrical AC power, which may be bidirectional. Data connectors 26 connected to data connector receptacles 27 transmit DC power from the backbone, i.e., the assembled frame without modules inserted, to modules 1, 12 on the one hand, and are configured for bidirectional data communication with the modules on the other hand.
[0073] The modules, i.e., PDU module 1 and / or gateway module 12, are removably attached to the frame via snap mechanisms that lock the modules in place once they are brought into position.
[0074] In one embodiment of the present invention, power distribution system 60 is configured to conserve energy consumed by dual PSU servers. This is achieved by performing the following method. In the first step, all PSUs are connected for a first predetermined time period. The first predetermined time period is preferably 10 to 50 minutes, more preferably 20 to 40 minutes, and most preferably 30 minutes. The controller 22 sends commands to the microcontrollers 17 in the different modules 1 to set all decoupling sections 16 of the phase switches 29 to the closed state. This is shown in Figure 10, where a dual PSU server is shown in four different decoupling switch 16 states. After step 1, the dual PSU server is in the state shown at the top of Figure 10, with the decoupling switches on both sides in the ON state. In a second step, power consumption measurements are made for each outlet and the power signals of each outlet, and therefore also for each PSU, are received by the microcontroller 17 in the PDU module. In the third step, the microcontroller 17 performs a health check on all PSU power signals by applying power quality measurement definitions from a standard such as, but not limited to, IEEE 1159-1995 to all PSU signals. This health check is performed if all PSUs are classified as "healthy" since disconnecting a PSU will not compromise server power requirements. A PSU is assumed healthy by registering its signal when entering first operation, and this is confirmed sensorily by the operator. In the fourth step, one of the two PSUs in each dual-PSU server is turned off for a second predetermined time period. The second predetermined time period is preferably 15-30 hours, more preferably 20-25 hours, and most preferably 23 hours and 30 minutes. This is illustrated by the second state in Figure 10, where the right PSU is in the off state and the left PSU is in the on state. In the fifth step, after a second predetermined time has elapsed, the first, second and third steps are performed again. This includes the third state of the dual PSU server shown in Figure 10, where both PSUs are again in an ON state. In the sixth step, the fourth step is performed again, but now the other of the two PSUs in each dual-PSU server is turned off for a second predetermined time. This is the fourth state in Figure 10, where the left PSU is in an off state and the right PSU is in an on state.
[0075] This method can be performed continuously until the health check determines an unhealthy PSU.
[0076] An advantage of controlling the power sent to the PSUs of a dual-PSU server according to the steps above is that it results in lower power consumption by each dual-PSU server. This is illustrated in FIG. 11, which shows the power consumption in both PSUs and the combined power consumption over a 48-hour period for an embodiment in which the first and second predetermined periods have the most favorable values, i.e., the first predetermined period is 30 minutes and the second predetermined period is 23 hours and 30 minutes. For the first 30 minutes, both PSUs are connected, and power consumption is at the level shown in segment 1P. After 30 minutes, power consumption has decreased, corresponding to the fourth step of the method described above. As shown in segment 2P, one of the two PSUs is disconnected, resulting in lower power consumption. After the second predetermined period of 23 hours and 30 minutes, both PSUs are reconnected to the power source for 30 minutes, so that both PSUs are reconnected to the power source, corresponding to the sixth step of the method described above. This is illustrated by segment 3P in FIG. 11. After this predetermined period of 30 minutes, segment 3P in Figure 11, thus a total of 24 hours and 30 minutes, the other of the two PSUs is now disconnected from the power source and power consumption drops again for 23 hours and 30 minutes, corresponding to the sixth step of the method described above. This is shown by segment 4P in Figure 11.
[0077] Figure 12 shows the efficiency of the PSU on the y-axis for each percentage of load on the x-axis. The graph shows that the efficiency of the PSU changes based on the percentage of load, and that increasing the load between 0% and approximately 50% increases the efficiency. In the above method shown in Figure 11, it can be seen that the multiple PSUs typically do not operate at loads greater than 50% but primarily operate between 10% and 30%. Therefore, the power supply that remains connected to the power source can be forced to a higher efficiency by carrying the load of the PSU that is disconnected from the power source. As shown in Figure 12, assuming that the load percentage of both PSUs is 15% when both PSUs are connected to the power source, disconnecting one PSU increases the load from 15% to 30% and increases the efficiency from 90% to 93%.
[0078] Furthermore, by implementing the above method, there is an added benefit that the circuits and fans of the disconnected PSU also do not consume energy.
[0079] This method can be extended to servers with more than two PSUs, where at least two PSUs can remain active, one on each power feed, thus maintaining redundancy unchanged.
[0080] 16, 17, 18 and 19 show a method for fault checking according to one embodiment of the present invention to prevent downtime due to emergency switch-on of disconnected PSUs based on monitoring of power quality phenomena.
[0081] FIG. 16 shows four dual-PSU servers 161, 162, 163, and 164, all connected to two power feeds. Thus, the four PSUs on the right are connected to Power Feed A, and the four PSUs on the left are connected to Power Feed B. For all outlets 6 connected to PSUs, the decoupling portions 16 of the phase switches 29 are closed; therefore, all PSUs are in the "on" state, as shown in FIG. 16. In this situation, the dual-PSU servers do not balance power consumption evenly across their PSUs. The right PSU is drawing 0.8 A, and the left PSU is drawing 0.2 A. As a result, one of the two PSUs in each server is not operating efficiently.
[0082] As shown in FIG. 17, in the same configuration of FIG. 16, some PSUs can be disconnected from the power feed by opening the decoupling sections 16 of some phase switches 29. When the decoupling sections 16 are open and the PSUs are therefore not connected to the power feed, the “off” state is shown in FIG. 17. For dual-PSU server 161, left PSU 175 is set to the “off” state, and right PSU 171 remains in the “on” state. For server 162, left PSU 176 is on and right PSU 172 is off. For server 163, left PSU 177 is off and right PSU 173 is on. For server 164, left PSU 178 is on and right PSU is off. Disconnecting one of the two PSUs in a dual-PSU server increases the power consumption of the PSU that remains connected to 0.95 A. As shown in Figure 17, by splitting the on PSUs between two power feeds, A and B, there is a total consumption of 1.9A across both power feeds. Comparing this to the situation in Figure 16 where all PSUs are on, a 5% reduction in energy consumption is achieved by disconnecting one of the two PSUs in a dual-PSU server.
[0083] However, operating a dual-PSU server with only one of the two power sources connected runs the risk of the server having no power if one of the two power sources fails. Figure 18 shows the situation in Figure 17 when power source B fails. In that situation, servers 162 and 164 are no longer connected to a power source.
[0084] To avoid the situation in Figure 18, in one embodiment of the present invention, power quality measurements are performed so that an emergency switch-on of PSUs 172 and 174 can be performed by closing the decoupling portions 16 of the phase switches 29 in the corresponding outlets 6, as shown in Figure 19.
[0085] For power quality measurements, the power quality definition of IEEE 1159-1995 may be used. [Table 1]
[0086] In one embodiment of the present invention, triggers can be created by applying the IEEE definitions for power quality measurements to current, total harmonic distortion, and power factor. These quality measurements are performed by the microcontroller 17 in the PDU module. When the microcontroller 17 determines a trigger, it instructs the "healthy" power source outlets to close all decoupling for PSUs that are in the off state. As a result, the server encounters the situation shown in Figure 19, where all PSUs are turned on in a healthy power source before the power source fails. A trigger could be high total harmonic distortion. Another trigger could be current expansion, sagging, or interruption. Yet another trigger could be a power factor drop. Another trigger could be detecting a deviation in the expected zero-crossing time of the voltage. This is shown in Figure 20. Figure 20 shows an AC voltage curve over time. The AC voltage curve has an expected zero-crossing 201 every 10 ms at 50 Hz or every 8.33 ms at 60 Hz, with very low jitter. In Figure 20, zero crossing 201 is a zero crossing during normal operation. A zero crossing may be detected later in time than expected, as indicated by zero crossing 202. A zero crossing may be detected earlier than expected, as indicated by zero crossing 203. Both late zero crossing 202 or early zero crossing 203 are triggers for a power loss. When such a trigger is detected, a command is sent to the outlet controllers of the alternate healthy power source to set all outlets to an ON state. This prevents the multi-feed server from being disabled when one of the two PSUs is disconnected from the power source and operating in an energy-saving mode.
Claims
1. 1. A computer-implemented method for distributing three-phase power in a rack system, comprising: the rack system includes a power distribution system for supplying power to electrical equipment, and is configured to be supplied with a three-phase power source; the power distribution system comprises at least one power bar with at least three outlets, a power sensor for measuring voltage and current at each outlet, and a phase switch for switching the live wires of the outlets; The method comprises: - receiving voltage and current values for each outlet; - calculating the power of each outlet based on the received voltage and current values; - registering the calculated power and corresponding phase setting for each outlet; - calculating a power load imbalance based on the calculated power and the phase settings of each outlet to obtain a calculated load imbalance; - calculating individual power calculations at each outlet via an iterative process, wherein the load imbalance is calculated for hypothetical combinations of phase configurations; - selecting the hypothetical combination with the lowest load imbalance as the best case; - comparing the best case load imbalance with the calculated load imbalance; if the best case load imbalance is lower than the calculated load imbalance, sending an instruction to the phase switch to set the phase setting to the best case phase setting; 11. A computer-implemented method comprising:
2. 10. A power distribution system comprising a processor configured to perform the method of claim 1.
3. A computer program comprising instructions which, when executed by a computer, cause the computer to carry out the steps of the method of claim 1.
4. 1. A computer-implemented method for distributing power in a rack system, comprising: the rack system includes a power distribution system for powering multi-PSU electrical equipment, the power distribution system being supplied with at least two power sources; The power distribution system comprises at least two power bars each having at least one outlet, a power sensor for measuring voltage and current signals of each outlet, and a decoupling switch for switching on or off live wires to the outlet; The method comprises: - receiving said voltage and current signals for each outlet; - applying power quality measurements to the received signals to determine a trigger for a power outage; If a power outage trigger is determined for an outlet connected to a multi-PSU device, determining another outlet connected to the multi-PSU device; sending a command to the other outlet to switch the decoupling switch to the on state; 11. A computer-implemented method comprising:
5. A power distribution system comprising a processor configured to perform the method of claim 4.
6. A computer program comprising instructions which, when executed by a computer, cause the computer to carry out the steps of the method of claim 4.
7. 1. A power distribution unit (PDU) module for distributing power to electrical equipment, such as servers and network switches, located in a data center rack, comprising: an input coupled to a power source and configured to receive three-phase power from said power source; a plurality of power lines coupled to the input, the power lines being configured to carry the three-phase power and including three live conductors and one neutral conductor; a plurality of outlets each coupled to at least two of the plurality of power lines to receive power from one of three phases of the three-phase power by connecting to one of the three live lines and the neutral line, each outlet configured to be coupled to an electrical device to power the electrical device, each outlet being provided with at least one power sensor configured to sense a parameter of the electrical power supplied to the at least one outlet; Equipped with the PDU module is configured to be connected to the gateway controller such that the gateway controller is connected to the at least one power sensor; the PDU module further comprises at least one three-phase switch and at least one decoupling switch, the at least one three-phase switch configured to selectively couple a phase contact of each outlet to one of the three live conductors; the at least one decoupling switch is configured to selectively disconnect the phase contacts of each outlet from the live conductor; Power Distribution Unit (PDU) module.
8. 8. The PDU module of claim 7, wherein one three-phase switch is provided for each outlet to select the live conductor connected to each outlet.
9. 8. The PDU module of claim 7, wherein one three-phase switch is provided for a group of outlets to select the live conductors connected to the group of outlets.
10. A PDU module according to any one of claims 7 to 9, wherein one decoupling switch is provided for each outlet to disconnect each outlet from the live conductor.
11. the parameter measured by the power sensor of the outlet is indicative of the load on the outlet, also referred to as a load parameter; The load parameter is preferably the current drawn by the outlet; the controller is configured to monitor the load parameters of the outlets in the PDU module and is further configured to generate instructions for operating the at least one three-phase switch and the at least one decoupling switch based on the monitored load parameters; The PDU module of any one of claims 7 to 10, wherein the PDU module is configured to receive the instructions from the gateway controller to control the three-phase switch and the decoupling switch.
12. the controller is configured to determine a total load for each phase based on the load parameters; the controller is further configured to generate instructions for operating the at least one three-phase switch and the at least one decoupling switch to increase load balancing of the three phases; The PDU module of claim 11 , wherein the PDU module is configured to receive the instructions from the gateway controller to control the three-phase switch and the decoupling switch.
13. the controller is configured to maintain a list of outlets, the phases delivered to the outlets, and the load parameters of the outlets; the controller is further configured to use the list to determine which outlets should switch phase; The PDU module of any one of claims 7 to 12, wherein the controller is configured to generate instructions for operating the three-phase switches accordingly.
14. 14. The PDU module of claim 7, wherein the controller is configured to determine a highest load phase and a lower load phase and to use the list to determine which outlets should switch from the highest load phase to one of the lower load phases.
15. The PDU module of any one of claims 7 to 14, wherein the list further comprises a priority of the outlet indicating the importance of reliable operation of the outlet.
16. 16. The PDU module of claim 7, wherein the controller is configured to receive, via a data connector, phase load information regarding the loads of the three phases of at least one external device coupled to the power source, and further configured to operate the at least one phase switch further based on the received phase load information.
17. the power sensor is enabled to sample the parameter at the microsecond level; 17. A PDU module according to any one of claims 7 to 16, wherein the controller is adapted to enable oscilloscope diagram graphical representation of the parameter in the time and / or frequency domain including representation of harmonics of the parameter.
18. The PDU module of any one of claims 7 to 17, wherein the controller is configured to identify anomalies in the time domain and / or frequency domain representation of the parameter.
19. 19. The PDU module of claim 7, wherein the controller is configured to compare the measured parameters with reference data retrieved from a cloud in which the measured parameters of the PDU module, preferably the measured parameters from other PDU modules, are stored, so as to enable identifying anomalies in the time domain and / or frequency domain representation of the parameters based on machine learning.
20. the parameter measured by the power sensor is the current delivered to the output; the input of the PDU module further comprises a voltage sensor configured to measure voltage between the power lines at the microsecond level; 20. The PDU module of any one of claims 7 to 19, wherein the controller is adapted to enable an oscilloscope diagram graphical representation of the voltage in the time and / or frequency domain including a representation of harmonics of the parameter.
21. 1. A power bar comprising an elongated frame configured to be mounted in a data center rack in a vertical 0U configuration or a horizontal 1U configuration, A power bar further comprising at least one PDU module according to any one of claims 7 to 21 attached to the frame.
22. 22. The power bar of claim 21, wherein said at least one PDU module is removably attached to said frame, thereby forming a modular power bar.
23. the power bar further comprises at least one network switch module having a plurality of network ports; The power bar of any one of claims 21 to 22, wherein the at least one network switch module is removably mounted to the frame.
24. The power bar of any one of claims 21 to 23, wherein the power bar comprises a plurality of PDU modules.
25. The power bar of any one of claims 21 to 24, wherein the controller is attached to the power bar.
26. the frame includes a power inlet connected to the power source; The power bar of any one of claims 21 to 25, wherein the modules mounted on the frame receive power from the power inlets.
27. the frame includes a power bus; The power bar of any one of claims 21 to 26, wherein the modules mounted on the frame connect to the power bus to transfer power from the power bus to the modules.
28. A first power bar according to any one of claims 21 to 27 configured to couple multi-PSU ICT equipment to a first three-phase power source, and a second power bar according to any one of claims 21 to 27 configured to couple the same multi-PSU ICT equipment to a second three-phase power source; the multi-PSU ICT device has two physically distinct power inlets; A power distribution system for powering multi-PSU ICT equipment, wherein a first power inlet is connected to the first three-phase power source and a second power inlet is connected to the second three-phase power source.
29. 30. The power distribution system of claim 28, wherein the system includes a gateway controller configured to switch power supply to multi-PSU ICT equipment using outlets provided in the first power bar and outlets provided in the second power bar.
30. 30. The power distribution system of claim 29, wherein the switching includes splitting power delivery to the multi-PSU ICT equipment between the outlets provided in the first power bar and the outlets provided in the second power bar according to a splitting scheme.
31. The division method is powering said electrical equipment entirely through the outlets provided on one of said first power bar and said second power bar, referred to as primary outlets; and upon detecting a failure of the primary outlet, switching to powering the electrical equipment entirely by the outlets of the other of the first and second power bars, referred to as a backup outlet.
32. all the outlets of the first power bar are primary outlets; 32. The power distribution system of claim 31 , wherein all the outlets of the second power bar are backup outlets.
33. 33. A power distribution system as claimed in any one of claims 28 to 32, wherein the first power bar and the second power bar are directly connected by a data connector, preferably a data cable, so as to increase the speed at which the controller can switch the outlets supplying power to the ICT equipment.
34. 34. The power distribution system of claim 33, wherein the gateway controller is provided in one of the first power bar or the second power bar.