AC vs DC power in data center

There has been a debate going on for some years if the traditional AC or new DC power distribution is best approach to power a data center. The DC power side has been pushing their technology with claims of quite considerable power savings. In many published articles, expected improvements of 10% to 30% in efficiency have been claimed for DC over AC. I have had my doubts of the numbers on their promises. Now there is some new data on AC vs DC issue available.

White paper compares AC vs. DC power distribution for data centers article tells that a new white paper from APC by Schneider Electric provides a quantitative comparison of high efficiency AC vs. DC power distribution platforms for data centers. The latest high efficiency AC and DC power distribution architectures are shown by the analysis to have virtually the same efficiency, suggesting that a move to a DC-based architecture is unwarranted on the basis of efficiency

A Quantitative Comparison of High Efficiency AC vs. DC Power Distribution for Data Centers paper demonstrates that the best AC power distribution systems today already achieve essentially the same efficiency as hypothetical future DC systems. It also tells that most of the quoted efficiency gains in the popular press are misleading, inaccurate, or false (like I have suspected to be for some time). And unlike virtually all other articles and papers on this subject, this paper includes citations and references for all of the quantitative data (which is very good).

The paper first describes that there are five methods of power distribution that can be realistically used in data centers: two basic types of alternating current (AC) power distribution and three basic types of direct current (DC) power distribution. These five types are explained and analyzed.

One AC and one DC, offer superior electrical efficiency. The paper focuses on comparing only those two highest efficiency distribution methods, which are very likely to become the preferred method for distributing power in future data centers. The data in this paper demonstrates that the best AC power distribution systems today already achieve essentially the same efficiency as hypothetical future DC systems.

The best AC system is based on the existing predominant 400/230 V AC distribution system currently used in virtually all data centers outside of North America and Japan. Increasing Data Center Efficiency by Using Improved High Density Power Distribution white paper gives details how it could be used in USA. It says that the use of the international 230/400 V distribution system instead of the USA standard 120/208 system can save 56% in the lifetime cost of the distribution system, and save floor space and weight loading.

The preferred DC system is based on a conceptual 380 V DC distribution system (consensus in the literature as a preferred standard) supplying IT equipment that has been modified to accept DC power. In the proposed international ETSI standard for DC distribution for data centers, the 380V DC system is actually created with the midpoint at ground potential to keep the maximum system voltage to ground to within +/- 190 V.

Based on the data I think the 400/230 V AC distribution system is the best way to go in data centers around the world.


  1. Tomi Engdahl says:

    High-voltage relays target electric vehicles

    Fujitsu’s FTR-E1 series of DC switching relays offers 13% higher voltage switching, 77% less power consumption, and a package size that is 40% smaller than similar relays. Intended for the electric vehicle market, these board-mount single-pole relays come with 12 VDC or 24 VDC coils.

    The FTR-E1 relays furnish contact ratings of up to 30 A at 450 VDC (40 A maximum/1 hour) resistive and consume just 900 mW at the rated coil voltage. Their compact package is 43.6×28.3×36.8 mm and weighs approximately 75 g, nearly 60% lighter than competitive relays, according to the manufacturer. In addition, the parts achieve an electrical life of 10,000 operations at 20 A, 450 VDC resistive.

    Devices provide a dielectric strength of 5000 VAC (1 minute) between coil and contacts and 2500 VDC (1 minute) between open contacts.

    Prices for the FTR-E1 series relays start at $29.85 each for 1 to 99 units.

  2. Pathak says:

    The utilization of DC for server farm or system room control genuinely constrains the sorts of IT gear that can be utilized. By and large operation is not down to earth without including a supplementary AC control framework. Reliable quality examinations amongst AC and DC control frameworks are profoundly subject to the presumptions made. A DC control framework is developed of a variety of DC rectifiers providing at least one parallel battery strings. Various late UPS item presentations use a comparable design, with a variety of UPS modules associated with a parallel exhibit of battery strings. Because of their closeness, DC and AC frameworks utilizing these outlines can be straightforwardly looked at.

  3. Tomi Engdahl says:

    Converters target microgrids, data centers–data-centers

    Industrial DC/DC converters from Powerbox operate with an input voltage of 180 V to 425 V and deliver 150 W, 300 W, 600 W, or 750 W of output power for use in microgrids and data centers. The PQB150-300S, PHB300-300S, PFB600-300S, and PFB750-300S series feature a layout optimized for thermal conduction, input/output isolation of 3000 VAC minimum, and single output voltages ranging from 3.3 V to 48 V.

    In lots of 1000 units, the PQB150-300S costs $95, the PHB300-300S costs $140, and the PFB750-300S costs $190.

  4. Tomi Engdahl says:

    How about power in the gas form to the server racks?

    Microsoft makes a ‘crazy’ bet on fuel cells to feed power-hungry data centers

    The tech company is testing the use of natural gas-powered fuel cells that could someday allow data centers — which consume 2 percent of U.S. electricity — to unplug from the power grid. That could translate into big cost savings and, potentially, cuts in carbon emissions.

    Twenty racks of servers sit in a stark, white, well-lit room — a familiar setup for anyone who’s visited one of the data centers that make up the humming infrastructure powering the internet.

    To see what’s special about this one, look up: Sitting on a steel frame above each stack of computer hardware is an electrical cabinet the size of a mini-fridge. Inside is a natural-gas-powered fuel cell.

    That technology, Microsoft engineer Sean James says, could allow future data centers to someday unplug from the power grid entirely.

    By generating electricity close by — literally on top of the computing hardware — Microsoft’s new design eliminates the inefficiency of producing electricity at a distant power plant and transporting it long distancesto data centers. That could trim the energy footprint of the fast-growing data-center business, eliminating a portion of the carbon emissions that fuel global warming, and, in the process, save Microsoft a lot of cash.

    The company’s Seattle trial is preliminary. But if Microsoft’s estimates hold up — and, a big if, the cost of fuel cells comes down — the savings of a fuel-cell-based design spread across the company’s fleet of facilities could total hundreds of millions of dollars.

    James sums up the prevailing view of the plan among the rest of the industry, a group that includes many conservative engineers content to tweak existing designs on the margins: “They think I’m crazy.”

    With demand for those services surging along with high-speed internet use, web giants Amazon, Microsoft and Google, as well as specialists like Digital Realty and Equinix, are scrambling to build warehouse-size data centers across the globe.

    That business is a massive, and growing, consumer of energy.

    Data centers account for about 2 percent of U.S. electricity use, the Department of Energy’s Lawrence Berkeley National Laboratory estimates, up from 0.8 percent in 2000. To cut their costs, companies like Microsoft have designed their newer facilities with energy efficiency in mind. They’ve also reduced their dependence on fossil fuels by buying renewable energy or building their own wind or solar farms.

    But Lucas Beran, who tracks data-center energy economics for IHS Markit, says the industry’s efficiency improvements have started to stall.

    “In the next few years we’re going to be at a crossroads,” he says. “We’ll have to change what we’re doing to maintain those energy gains.”

    Energy experiments

    Microsoft’s fuel-cell concept stems from years of experimentation.

    Mock data center

    Microsoft isn’t the only technology giant dabbling in fuel cells.

    Apple and eBay have used fuel cells to power data centers from a centralized location, essentially replacing the backup generators or grid connections in a typical data-center design with fuel-cell clusters.

    With that clearance,McKinstry, the Seattle-based contractor that built and is hosting Microsoft’s experiment in a formerly vacant space attached to its headquarters, will link the mock data center to the municipal natural-gas grid.

    Gas will be piped to the 20 fuel cells, starting an electrochemical reaction that extracts hydrogen atoms and sends a current of negatively charged electrons to power the servers below.

    Waste products — water vapor and a small amount of carbon dioxide — will be vented out of the building along with the excess heat from the servers.

    In a real data center, the servers would be processing Bing web searches or storing customers’ email. For the purposes of the trial, the 20 racks in Sodo will be filled with dummy data meant to simulate actual workload.

    Microsoft will add methane detectors to guard against potential gas leaks, and airflow monitoring to see how the design deals with exhaust.

    Microsoft researchers, in tests a few years ago with the University of California, Irvine, estimated that when plugged into the power grid, the average data center reaped about 17 percent of the potential energy of the fuel used to generate that electricity.

    The in-rack fuel-cell concept can pull off 29 percent efficiency, Microsoft estimates, because no energy is lost through the long haul from power plant to conversion and consumption, and because the fuel cell’s chemical reaction is more efficient than some industrial-scale power generation.

    Pros and cons

    There is a problem, though. Fuel cells are expensive. Current models cost about twice as much as Microsoft needs to make the concept pay off.

    But the company is optimistic. Fuel-cell manufacturing is a relatively new industry, with most fuel cells bound for relatively niche applications like backup power, cranes and industrial equipment, and specialty vehicles. If big buyers such as Microsoft start lining up for many thousands of them, their costs may come down.

    In that case, the savings would be significant. Microsoft researchers estimate that mass-produced fuel cells would cut the cost of installing a new data-center rack by at least 10 percent, and the costs of operating that rack by 21 percent.

    Those savings pencil out, conservatively, to about $80 per rack, per month. With more than 1 million servers in Microsoft’s worldwide data-center fleet, the potential savings could stretch into the hundreds of millions of dollars a year if the design were rolled out across the board.

  5. Tomi Engdahl says:

    Power Transformers Aim for a Bigger Role in the Smart Grid

    Improvements in power transformers will address the needs of the smart grid. Benefits include enhanced energy security, reduced greenhouse gas emissions, improved urban air quality, and greater grid utilization.

    This is particularly true on the distribution side of electric power, where the total cost of ownership and payback for a utility and the end user over the useful life of a transformer is taken into consideration. According to Metglas Inc., a fully owned subsidiary of Hitachi America Metals Ltd., amorphous metal power transformers are less costly and more efficient than cold-rolled grain-oriented steel transformers.

    The company manufactures amorphous metal power transformers using a proprietary process. The key to Metglas’ rapid-solidification manufacturing process is cooling the molten alloy at a rate of approximately 1 million ºC/s, using a melt spinning technique.

    Metglas says hundreds of utilities are benefiting from amorphous-metal core transformers. By replacing conventional distribution transformers with amorphous-metal types, approximately 27 TWh of core losses in the U.S. alone could result in annual savings.

    Amorphous core technology is also used by the United Kingdom’s Wilson Power Solutions, which is upgrading its products for the smart grid. Its e2+ transformers come with a 17-position on-load tap changer that adjusts the taps automatically to maintain a constant secondary output voltage (relative to input voltage fluctuations). This is done via a volt-ampere reactive relay, where site supply voltages fluctuate, or a constant ±1% output voltage is required.

    Despite the fact that most power transformers are not quite yet ready for the emerging smart grid, there’s room for improvement for incorporating remote diagnostic smart monitoring of a wide range of transformer and system parameters

    Dickinson sees transformers used in power transmission as immediate candidates for integration in the smart grid, with resulting immediate benefits of enhanced energy security, reduced greenhouse gas emissions, improved urban air quality and greater grid utilization.

  6. Tomi Engdahl says:

    Technology Award for Better Electricity Delivery’

    The innovation developed by the Lappeenranta University of Technology “Low Voltage Equal Distribution System for Public Power Distribution” was awarded the Technology Prize by the Finnish Messrs Foundation. The prize for the first time is € 10,000. The award was announced today at the Technology 17 event opened at the Helsinki Fair Center.

    Since 2005, Lappeenranta University of Technology has been researching a power distribution system based on low-voltage electricity (LVDC) in cooperation with industry. The aim has been to develop a more cost-effective way of replacing existing electricity distribution systems with renewable electricity distribution networks to meet future demands.

    - As a result of the research, there is an innovation from active LVDC distribution, where the familiar energy of the history is connected to modern power electronics, network technology and information systems. LVDC’s business potential in Finland is € 40-60 million a year and its estimated global annual market potential is billions in billions, says Professor Jarmo Partanen.

    - Innovation makes it possible to build electricity distribution networks at a lower cost, to improve the reliability of electricity distribution and the lack of security. In addition, similar technologies have many other applications, ships, real estate, electric car charging systems, said sales manager Marcus Bergström from the Exhibition Center as secretary of the prize drawer.


  7. Tomi Engdahl says:

    Data center power market charted for steady increase til 2024: Report

    Key players in the global data center power market, according to the report, include Server Technology Inc., Schneider Electric SE, Rittal GmbH & Co. KG, Intel Corporation, Emerson Network Power, Raritan Inc., General Electric Company, Hewlett-Packard Company, and Eaton Corporation Plc/ABB Ltd.

    Data Center Power Market to Witness Steady Increase by 2024

    Zion Market Research, the market research group announced the analysis report titled “Data Center Power Market: Global Industry Analysis, Size, Share, Growth, Trends, and Forecasts 2016–2024″

  8. Tomi Engdahl says:

    Are Solid-State Transformers Ready for Prime Time?

    It is possible that “solid-state transformers” could reduce the size and weight of power distribution systems, but we’re not quite there yet.

    Several companies are working on technologies that could replace large traditional power transformers with power semiconductors and smaller transformers mounted on circuit boards. Although they are called solid-state transformers, they are really power converters.

    The figure below is a conceptual circuit for a “solid-state transformer” that accepts a three-phase 60 Hz high-voltage input and provides a 60 Hz lower-voltage output. The transistors could be SiC or GaN types with the appropriate specifications. The input circuit converts 60 Hz high voltage ac input to a dc voltage. Then, the dc produces an ac voltage of 10 to 20 kHz that is applied to a step-down transformer. The transformer output is converted to dc and applied to an inverter to produce a lower voltage 60 Hz ac output. The transformer is necessary to provide isolation between the input and output. An advantage of this approach is reduction in size and weight of the transformer because it can operate at a much higher frequency than a 60Hz power transformer.

    Microgrids could be deployed much more rapidly. Grid efficiency could conceivably be increased by up to 8% to 10% because of lower conversion and transmission losses.”

    In describing the advantage of this approach, “that’s 10% less power that you have to generate,” says John Palmour, co-founder and chief technology officer for Power and RF at Cree, a producer of SiCs. “We can replace an 8,000-pound transformer in a substation running at 60 Hz and replace it with one running at 20 kHz in a tiny design. We can shrink it down to [the size of] a suitcase.”

    NC State Report Says Solid-State Transformers Are Ready

    Now, fast foward to July 2017. A North Carolina State (NC State) study using complex computational models found that smart solid-state transformers (SSTs) could be used to let power distribution systems route renewable energy from homes and businesses into the grid. Such a grid would improve efficient use of renewable energy and storage but, to date, this version of the smart grid has been mostly conceptual, the report added.

    “Using this model, we found that SSTs can greatly enhance the functionalities of tomorrow’s power grid,” Chakrabortty says. “However, certain operational boundaries would need to be maintained.”

    Essentially, system designers and operators would need to ensure the system—at every level —is taking into account customer power demand, power generation from renewable sources, and energy storage capacity, in order to avoid providing too much or too little power.

  9. Tomi Engdahl says:

    The Data Center’s backup power to the grid reserve

    Eaton is developing, for example, a service center that provides them with the opportunity to participate in the frequency control of the network with uninterruptible power supplies, ie UPSs. This UPS-as-a-Reserve (UPSaaR) service is the first of its kind in the data center.

    The service provides data center operators with the ability to keep their system within the limits of the power grid and thus avoids network-wide power outages by quickly adjusting the current power consumption. The service is targeted at large data center operators such as cloud service providers and will be brought to the European market by the end of 2017.

    Eaton has developed a service in close cooperation with Fortum. As energy markets move from fuel-based products to renewable energy, energy production itself may become more volatile and the anticipation and balancing of power generation will become more difficult. Network operators buy from service providers, such as energy producers, frequency-controlled operating and disturbance resources to balance electricity consumption and production to maintain frequency.

    Comprehensive tests between Eaton and Fortum have demonstrated that uninterruptible power supply systems and batteries can be safely and efficiently used to perform supply-demand functions without the risk of uninterrupted power supply main operation. Uninterruptible power supply (UPS) can function as part of a virtual power plant, enabling data centers to participate in highly money-intensive frequency-controlled operating or disturbance and demand-driven markets.


    Eaton UPS-as-a-Reserve Solution

  10. Tomi Engdahl says:

    AC versus DC load breaking comparison with a knife switch
    AC Circuit Breaker 230V DC Test

  11. Tomi Engdahl says:

    Why Do Today’s Server Applications Use 54-V BLDC Motors?

    More server manufacturers are adopting 54-V brushless dc motors over traditional 12-V BLDCs to achieve significant savings on a couple of fronts.

    Cloud-based computing refers to a mesh of remote servers that stores and moves data around the world so that we can access via Wi-Fi, local-area network (LAN), or a cellular network.

    These remote servers act as a large storage device that consists of clusters of servers in a warehouse commonly referred to as a server farm. These server farms require a constant ambient temperature (optimal temperature range is between 68° and 71°F) to operate at their highest performance and to minimize any failure. They’re typically cooled by central air conditioning or heated with central heating depending on their location, just like a typical office space.

    Traditionally, server applications have used 12-V BLDC fans to cool the electronics in a cabinet. However, just like automotive applications, 54-V BLDC motors are being adapted for server applications for several reasons.

    Server manufacturers are adopting 54-V BLDC motors over traditional 12-V BLDC motors because it allows them to use one fourth of the current. In turn, motor manufacturers can use thinner ­­copper wire. This also enables motor manufacturers to reduce the size of the motor and, therefore, the overall cost of the motor

    For example, in a 450-W server, 32 W are consumed by the 12-V BLDC fans.

    One issue does emerge when dealing with the electronics that drive a 54-V BLDC fan motor: Server engineers can’t use the old 12-V hardware to drive 54-V motors. They’re required to use electronic components with a higher operating voltage that are suitable for a 54-V power supply with plenty of margin.

    Nonetheless, several hardware solutions on the market can help ease this transition.

    The MIC28514 converts the 54-V supply bus rail to a traditional 12-V power rail with better than 90% power efficiency. As a result, server engineers can continue to use the same motor-control algorithms and proven active components.

    These high-voltage devices make it feasible for server manufacturers to adopt 54-V power bus technology and, in turn, reduce overall system cost by utilizing smaller motors and less copper width on PCB boards and cabling.

  12. Tomi Engdahl says:

    Low-Voltage Logic Inputs Turn 1000Vdc Loads On/Off

    Computer logic controls up to a 1000Vdc load using a 22mm x 9mm x 5.16mm BGA micromodule employing an internal isolated Power Switch Controller that drives an external power MOSFET or IGBT.

    The LTM9100 µModule (micromodule) from Linear Technology accepts logic inputs that enable its internal isolated Power Switch Controller to drive an external-power MOSFET/IGBT switching at up to1000Vdc. It uses a galvanic isolation barrier to separate logic inputs from its Power Switch Controller, which can turn high-voltage power sources on and off. In doing so, the isolation barrier protects its low-voltage logic inputs from the neighboring high-voltage Power Switch Controller.

    Many computer-based applications employ high voltages that you can control with the LTM9100. One such application is industrial motor drives that may operate from 170Vdc to 680Vdc. Grid-tied solar systems can operate up to 600V or more. Primary power for some modern fighter aircraft is 270Vdc. Li-ion batteries in electric vehicles can reach up to 400V.

    The key to the LTM9100’s power protection is its internal 5kVRMS galvanic isolation barrier that separates the digital input interface from the Power Switch Controller that drives an external N-channel MOSFET or IGBT switch

    You can configure this isolated power switch controller for use in either high side or low side applications (hence, its Anyside name), as shown in Fig. 2. In addition, it can be used in floating applications.

    Adjustable undervoltage and overvoltage lockout thresholds ensure that the load operates only when the input supply is in its valid range. A current-limited circuit breaker protects the supply from overload and short-circuits.

    This isolated power switch controller minimizes inrush current by soft-starting the load. It is versatile enough to control inrush current in hot-swappable cards, ac transformers, motor drives, and inductive loads.

  13. Tomi Engdahl says:

    Besides data centers high voltage DC power is used on modern cars at approximately same voltage levels:

    Driving the Future of HEV/EV with High-Voltage Solutions

    Energy efficiency has become a key global focus because of its contribution toward reduced carbon-dioxide emissions. One significant area of contribution is the electrification of vehicular technology.

    hybrid electric vehicles and electric vehicles (HEVs/EVs). HEV/EV sales are expected to represent between 5 and 20% of all cars sold by 2025

    The foundation for HEV/EV architectures is high voltage. These vehicles are based on high-voltage battery systems, such as +400 V for EVs and 48V for HEVs.

    SMPS conditioners are realized in these power train sub-systems in HEVs/EVs:
    48 V to 12 V bidirectional power supplies
    400 V batteries (EVs only)
    Bidirectional 400 V to 12 V power supplies

  14. Tomi Engdahl says:

    Are Solid-State Transformers Ready for Prime Time?
    It is possible that “solid-state transformers” could reduce the size and weight of power distribution systems, but we’re not quite there yet.

  15. Tomi Engdahl says:

    Transformers: ensuring efficient electrical system design

    Electrical engineers can specify high-efficiency, 3-phase, low-voltage, dry-type transformers into facilities, taking into consideration the many parameters of transformer design to create a safe, effective electrical distribution system.

    A key component of power delivery and distribution is the transformer, which is an apparatus for reducing or increasing the voltage of an ac system. It is easier to transmit power at a higher voltage and a lower current. However, the higher voltage is unusable in daily operations. Therefore, it is necessary to transform it to a more usable voltage level for safe operation. It is difficult to get power across long distances at low voltage resulting in a need for reasonably low voltages at points of use.

    Depending on where you live relative to a power plant, the number of transformers in the path of power delivery could be quite considerable. There could be a series of step-up, step-down, isolation, or voltage-correction transformers.

    There are many types of transformers today, but the most common type is dry-type. Most 3-phase transformers used in commercial and industrial applications are dry-type transformers. This article examines the use of 3-phase, dry-type transformers in facility design and some important things to consider when specifying high-efficiency transformers.

    Many associations and agencies have been involved in 3-phase transformer efficiency, which has allowed this development in the U.S. to advance considerably since the 1990s. Transformers were, and are still, considered necessary components of an electrical distribution system.

    Low-voltage, 3-phase, dry-type transformer efficiency has always been driven at the manufacturing level. It is up to the manufacturer’s designers to use materials that will meet the requirements. The 1996 NEMA TP-1 standard included requirements for 97% efficiency for smaller transformers (up to 15 kVA) and up to 98.9% efficiency for larger transformers (1,000 kVA or more). The current 2016 DOE requirements for transformer efficiency mandate 97.79% efficiency for smaller transformers and up to 99.28% efficiency for the larger type.

    Design considerations

    As electrical distribution system designs are using the new transformers, engineers and designers should consider physical space, rated temperature rise, impedance, maximum inrush current, and arc flash.

    Physical space. As energy efficiency has been the focus of transformer design, the physical footprint of the transformer has noticeably become larger to accommodate design revisions.

    Rated temperature rise. Standard temperature rise is the rated operating temperature above the ambient temperature. When one is converting power from higher to lower voltage or vice versa, heat is generated. Lower temperature rise is achieved by incorporating a more robust core-and-winding configuration. Standard transformer temperature rises are rated for 302°F, 239°F, and 176°F. The most commonly used temperature-rise rating is 302°F because it has the least amount of material in the core and windings, making it less expensive to manufacture and purchase.


    Impedance is the effective resistance of the electric circuit or component to ac, arising from the combined effects of ohmic resistance and reactants. As transformers become more efficient, the result is lower impedance. Some impedance is a useful concept when designing a distribution system. The impedance helps reduce the available fault current between points in the system. At different, specified, temperature-rise ratings, the impedance can vary for the same size transformer. A transformer rated for 112.5 kVA may have an infinite bus let-through greater than 10,000-ampere interrupting capacity (AIC) depending on the temperature rating specified, resulting in downstream equipment needing to be rated for 14,000 AIC or more.

    Maximum inrush

    Inrush is the sudden influx of power. So, with greater efficiency, modified construction, and lower impedance, the maximum inrush at 100% load can be higher than seen in previous, less efficient transformers.

    Arc flash considerations

    Arc flash is caused by available incident energy in an electrical distribution system. It is a type of electrical explosion or discharge that results from a low-impedance connection.

    NFPA 70E-2015: Standard for Electrical Safety in the Workplace allows for transformers sized for 240 V at 125 kVA and less to be omitted from incident-energy calculations because the authors determined that the equipment downstream of these transformers could not generate enough incident energy to be harmful.
    use of the calculated values for all transformer sizes below 125 kVA is recommended for the engineer’s consideration. This is a more conservative approach to arc flash analysis, but human safety is the highest priority.

    efficiency is the name of the game when it comes to designing on a budget, not just for the capital investment, but for the overall operational cost of the facility.

    Power-usage effectiveness (PUE) defines overall energy efficiency in a data center’s power distribution system. The equation is simple. It is all power consumed at the facility for the IT load, plus the mechanical and auxiliary systems required to operate the facility, divided by the IT load. PUE for the client’s older mission critical facilities was in the 2.0 range. This meant that it took twice as much energy to cool the IT equipment than it did to operate the equipment, and the IT load was equivalent to the HVAC load.

    To save energy, many facilities now use “free” cooling to prevent the mechanical cooling system from operating.

    In both facilities, the client wanted to achieve an annualized PUE of 1.2 or better

    The electrical distribution system within the first facility was there to provide a lot of power in a small space. The power density into the whitespace was designed to support a 12-kW IT cabinet. In the first facility, power was taken at 25 kV and converted at the utility level to 480 V via 18 2,500-kVA pad-mount transformers. It was then converted to 208/120 V inside the facility for IT equipment use. The 480 V-to-208/120 V conversion was done via 36 750-kVA transformers. Based on the manufacturer’s information, each of these transformers had a total loss of approximately 15,400 W, totaling approximately 554.4 kW of losses in the facility due to transformer inefficiency. Over a period of a year, this transformer loss resulted in approximately 4.86 million kWh of lost energy, for which the client had to pay just to operate.

    In the second facility, the request was to be more efficient than the first facility.

    Some utility providers will offer a lower cost-per-kWh rate if the client owns its own medium-voltage distribution. With this in mind, the client decided to take delivery at 480 V and not purchase its own medium-voltage equipment. This same approach was taken with the second facility. Instead of taking delivery at 480 V only, the utility provider allowed CRB to take delivery at multiple voltages. The second facility used pad-mounted transformers with 480 V secondary windings to support mechanical equipment within the facility and pad-mounted transformers at 208/120 V for connection directly to IT equipment. This mixed-voltage delivery approach minimized the number of transformers required inside the facility and removed the transformer losses

  16. Tomi Engdahl says:

    Power play: cut costs in backup battery systems

    As telecommunications providers respond to growing service demands with ever-increasing power density, a need has arisen for additional equipment to handle the load while reducing the facility’s total cost of ownership (TCO).

    One way of reducing TCO is to reduce the frequency of battery replacements and interruptions of service by using backup battery systems that are designed and manufactured to meet the demands of the application. These backup battery systems typically use lead acid batteries, which provide the power density that telecommunications networks require.

    A significant factor that affects the life of lead acid batteries is temperature. Lead acid batteries function best at an average of 77 degrees F (25 degrees C); high temperatures severely reduce life expectancy. For every 15 degrees above 77 degrees that a valve regulated lead acid (VRLA) battery experiences, the float life is reduced by half. The environment in which the batteries are used, such as in the open air, can subject them to high temperatures that compromise their service life.

    Option 1: Outdoor placement

    In order to make better use of indoor facility space in the face of additional equipment, some telecommunications facilities have chosen to “take it outside”; i.e., to move the backup battery system out of the building and into an outdoor enclosure designed for the purpose.

    Option 2: Better batteries

    Telecommunications facilities have an additional cost-reduction option in reserve power system selection: batteries that feature thin plate pure lead (TPPL) technology. Batteries with TPPL technology use 99.99% pure lead for the battery grids and active material. In addition, the high-purity electrolyte provides an extremely low float current and high recombination efficiency, which slows the drying out of the electrolyte and corrosion of the plates to extend service life.
    For applications in telecommunications infrastructure, TPPL batteries can withstand higher temperatures than conventional lead acid batteries.


    Telecommunications operators have options to reduce their TCO in the selection of backup battery systems. These options include housing batteries in thermally managed outdoor enclosures and using batteries with TPPL technology. The selection of the right thermal management solution and battery type to minimize TCO can depend on location.

  17. Tomi Engdahl says:

    Overhead Power Distribution. Made Simple.

    First introduced in 1987, Starline Track Busway is now the industry-leading electrical power distribution system for the data center/mission critical, retail, industrial, and higher education markets – or for any facility where flexible power is needed.

    Available in 40, 50, 60, 100, 225, 250, 400, 800 or 1200 amp systems

    Easiest installation of plug-in units—just insert and turn

  18. Tomi Engdahl says:

    Now there’s a better alternative to alternating current.

    The digital age has been with us for quite some time. But the technology needed to reliably and efficiently supply the direct current power that IT equipment needs has not been available. Until now. Starline DC Solutions enables data centers and other DC centric applications to adapt to current and future power generation systems (solar, wind, fuel cells, etc.) in the form of reliable 380V Direct Current. The managers, users, designers and those financially responsible for data centers are now focusing on reliability in power consumption support systems that are adaptable to new and evolving power generation techniques without affecting data center operations.

    Now considered to be large power consumers, data centers require power support systems with flexible modularity, higher efficiency, more cost effective reliability, and a smaller footprint while enjoying the security of global safety listings. Starline DC Solutions has developed the unique power support system technology that makes 380V DC power a reality now.

    Starline Track Busway offers a unique and proven solution for easy and flexible high-voltage DC distribution. With systems available with ratings up to 600V DC

    Several rectifiers are available from Starline DC Solutions for integration into your facility. With units ranging from 30kW to 120kW, we can accommodate whatever your power requirements are for deployment within your facility. A variety of battery storage arrays that pair directly with the rectifiers can also be purchased.

    Starline DC Solutions offers several products to assist with integration of your renewable energy sources into the facility.

  19. Tomi Engdahl says:

    Power Controller Merges, Manages Multiple Large-Scale Energy Sources

    A 60-kW demo system developed by NREL accepts solar, generator, and battery power, then combines and manages these sources for optimal efficiency and run time.

    There’s a management-and-control problem when multiple power sources—solar, diesel generator, and batteries—each has variable, non-static availability. Deciding which one(s) to use and to what extent must be done properly for maximum efficiency and performance, of course. It’s possible to manually switch among them, but that’s awkward, carries risk of misconnection, and is inefficient.

    Engineers at the Energy Department’s National Renewable Energy Laboratory (NREL), in conjunction with KBR (El Segundo, Calif., formerly Wyle Laboratories), worked to solve just such a problem for the U.S. Army’s field installations.

    That’s where NREL’s Consolidated Utility Base Energy—CUBE—could play a major role (Fig. 1). The unit, which measures about three feet wide, six feet long, and five feet tall, converts power from different sources (solar panels, batteries, and generators) into usable electricity, functioning as a microgrid. It quickly and autonomously switches among the different sources to deliver uninterrupted power.

    Though the CUBE doesn’t store energy, instead the unit manages its sourcing and flow

    The hybrid power system integrates four 5- to 10-kW photovoltaic (PV) arrays, one 30-kW battery pack, and two 30-kW diesel generator sets to support a 60-kW peak load

    The CUBE provides three key functions: power conversion, power distribution, and power protection (extremely critical in field installations)

    Nearly all of the hardware is standard, off-the-shelf components and modules. The control platform is based on National Instruments’ CompactRIO modules, each of which includes an FPGA backplane and a real-time processor, while the power-electronic converters use Semikron SKAI modules. These modules include insulated-gate bipolar transistor (IGBT) power semiconductors; dc link capacitors; gate drivers with low-level protection; and voltage, current, and temperature sensors.

    Applications for CUBE go beyond military situations. It could be used in other remote locations, or even provide emergency power management after natural disasters.

  20. Tomi Engdahl says:

    Data center design considerations

    This article provides guidelines on distribution systems’ levels of redundancy, the correct generator rating to use, and whether solar power can be used in a data center.

    Learning objectives:

    Know the difference between 2N, 3M2, and N+1 system topologies.
    Understand the characteristics of the system topologies.
    Learn the criteria for rating generators.
    Understand the different transformer types.

    What is the best system topology?

    There isn’t a single “best” system topology. There is only the best topology for an individual data center end user.

    2N: Simply designing twice as much equipment as needed for the base (i.e., N) load and using static transfer switches (STS), automatic transfer switches (ATS), and the information technology (IT) and HVAC equipment’s dual cording to transfer the load between systems. The systems are aligned in an “A/B” configuration and the load is divided evenly over the two systems. In the event of failure or maintenance of one system, the overall topology goes to an N level of redundancy.
    3M2: This topology aligns the load over more than two independent systems. The distributed redundant topology is commonly deployed in a “three-to-make-two” (3M2) configuration, which allows more of the capacity of the equipment to be used while maintaining sufficient redundancy for the load in the event of a failure (see Figure 2). The systems are aligned in an “A/B/C” configuration, where if one system fails (e.g., A), the other two (B and C) will accept and support the critical load. The load is evenly divided with each system supporting 33.4% of the load or up to 66.7% of the equipment rating. In the event of a component failure or maintenance in one system, the overall topology goes to an N level of redundancy. In theory, additional systems could be supplied, such as 4M3 or 5M4, but deployment can significantly complicate the load management and increases the probability of operator error.
    N+1 (SR): The shared-redundant (SR) topology concept defines critical-load blocks. Each block is supported 100% by its associated electrical system. In the event of maintenance or a failure, the unsupported equipment would be transferred to a backup system that can support one or two blocks depending on the design. This backup system is shared across multiple blocks, with the number of blocks supported being left to the design team but typically in the range of 4:1 up to 6:1.
    N+1 (CB): The common-bus (CB) redundant system is like the shared redundant system in that the IT equipment’s A and B sources are connected to an N+1 uninterruptible power supply (UPS) source, but in the event of a failure or maintenance activities, the load is transferred to a raw power source via STS. The raw power source has the capability of being backed up by generators that are required to be run during maintenance activities to maintain the critical load.

    The above topologies assume a low-voltage UPS installation. However, similar systems can be developed using a medium-voltage UPS.

    2N topology. The premise behind a 2N system is that there are two occurrences of each piece of critical electrical equipment to allow the failure or maintenance of any one piece without impacting the overall operation of the data center IT equipment.

    Distributed redundant (3M2) topology. The premise behind a 3M2 system is that there are three independent paths for power to flow, each path designed to run at approximately 66.7% of its rated capacity and at 100% during a failure or maintenance event.

    N+1 shared redundant (N+1 SR). The premise behind the N+1 SR system is that each IT block is supported by one primary path. In the event of maintenance or a failure, there is a redundant but shared module that provides backup support

    N+1 common bus (N+1 CB). The premise behind the N+1 CB system is there is one primary path that supports each IT block. This path also has an N+1 capacity UPS to facilitate maintenance and function in the event of a UPS failure. The system is backed up by a simple transfer switch system with a backup generator.

    What generator rating should be used for a data center?

    There are four ratings defined by ISO-8528:

    Continuous power: designed for a constant load and unlimited operating hours; provides 100% of the nameplate rating for 100% of the operating hours.
    Prime power: designed for a variable load and unlimited running hours; provides 100% of nameplate rating for a short period but with a load factor of 70%; 10% overload is allowed for a maximum of 1 hour in 12 hours and no more than 25 hours/year.
    Limited running: designed for a constant load with a maximum run time of 500 hours annually; same nameplate rating as a prime-rated unit but allows for a load factor of up to 100%; there is no allowance for a 10% overload.
    Emergency standby power: designed for a variable load with a maximum run time of 200 hours/year; rated to run at 70% of the nameplate.

    What IT distribution voltage should be used?

    By now it’s well understood in the data center industry that 3-phase circuits can provide more power to the IT cabinet than a single-phase circuit. However, the choice of distribution voltage between 208 Y/120 V or 415 Y/240 V depends on the answers to several questions, such as:

    How much power needs to be delivered to each IT cabinet initially, and what does the power-growth curve look like for the future?
    What are the requirements of the IT equipment power supplies?
    Will legacy equipment be installed in the data center?
    Can the facilities team decide on the power supplies to be ordered when new IT equipment is purchased?

    Let’s start with the power of a 3-phase circuit. A 208 Y/120 V, 3-phase, 20-amp circuit can power up to a 5.7-kVA cabinet. Per NEC Article 210.20, branch-circuit breakers can be used up to 80% of their rating, assuming it’s not a 100%-rated device. Therefore, a 208 V, 3-phase, 20-amp circuit can power a cabinet up to 5.7 kVA (20 amps x 0.8 x √3 x 208 V). Now, if that same 20-amp circuit was operating at 415 Y/240 V, 3-phase, then that circuit could power a cabinet up to 11.5 kVA (20 amps x 0.8 x √3 x 415 V). That’s more than twice the power from the same circuit for no extra distribution cost.

    If the specifications for the IT equipment can be tightly controlled, the decision to standardize on 415 Y/240 V distribution is a pretty simple one. However, if the IT environment cannot be tightly controlled, the decision is more challenging. Currently, most IT power supplies have a wide range of operating voltage, from 110 V to 240 V. This allows the equipment to be powered from numerous voltage options while only having to change the plug configuration to the power supply.

    However, legacy equipment or specialized IT equipment may have very precise voltage requirements, thereby not allowing for operation at the higher 240 V level. To address this problem, both 208 Y/120 V and 415 Y/120 V can be deployed within a data center, but this is rarely done as it creates confusion for deployment of IT equipment.

    The follow-on question typically asked is if the entire data center can run at 415 V, rather than bringing in 480 V and having the energy loss associated with the transformation to 415 V. While technically feasible, the equipment costs are high because standard HVAC motors operate at 480 V.

    Must we install an emergency power-off system?

    Emergency power-off (EPO) buttons are the fear of every data center operator. With the push of a button, the entire data center power and cooling can be shut down. Because of the devastation that activation of an EPO can cause, EPOs typically are designed with a two- or three-step activation process, such as lifting a cover and pressing the button or having two EPO buttons that must be activated simultaneously. These multistep options assume that the authority having jurisdiction has provided approval for such a design. However, EPOs are not necessarily required.

    Can we use photovoltaic systems to power our data center?

    Corporations and data center investors are demanding sustainability be built into the data center. The positive impact on public relations by showcasing a sustainable data center can’t be underestimated, especially considering how much of a power hog data centers can be.

    A PV system may or may not provide power during a utility power failure, depending on the type of inverter installed.

    The trend is to provide a PV system that offsets some of the noncritical-administration power usage. These systems are typically small (less than 500 kW) and can be located on building rooftops, carports, and on the ground.

    A medium-voltage alternative to low-voltage UPS

    Design topology evaluation also should consider the medium-voltage uninterruptible power supply (UPS). Like the topologies using the low-voltage UPS, the medium-voltage UPS can be deployed in 2N, N+1, and 3N/2 configurations. Regardless of the topology used, medium-voltage UPS systems offer advantages over low-voltage UPS systems. They generally are installed outdoors in containers, thereby minimizing the conditioned building footprint.

  21. Tomi Engdahl says:

    Data Center Circuit Protection Selection Guide

    To keep these critical facilities running properly, data center operators need advanced circuit protection, sensing, switching and power management components.

    The Data Center Circuit Protection Selection Guide simplifies identifying the right components for a variety of applications, including server protection, network equipment port protection, and UPS protection.

  22. Tomi Engdahl says:

    Choosing 600 or 1000 VDC in Photovoltaic Projects

    The photovoltaic (PV) industry continues to benefit from the lower costs of solar modules and advances in inverter technology. One of the more significant trends in recent years is the migration from 600 Volt DC systems to 1000 Volt DC systems. For several years, 1000 V systems have dominated the utility “behind the fence” market to reduce installation costs and improve performance. To achieve these same advantages, some commercial systems are now being installed with 1000 V strings. With proper planning and practices, 1000 V systems can be just as safe as 600 V systems.

    The NEC only prohibits PV systems over 600 V in one- and two-family dwellings. A dwelling for three or more families or any commercial site could, therefore, be a candidate for a 1000 V system.

    With systems over 600 V, there is a need to prohibit access by unqualified personnel, and there are
    also higher rating requirements for the DC equipment.

    Electrical equipment operating
    above 600 V must be housed within a metal
    enclosure that is marked with signs prohibiting
    access by unqualified persons.

    Conductors over 600 V
    cannot occupy the same equipment wiring
    enclosure, conduit, or raceway as conductors
    rated for less than 600 V, including Ethernet
    or other network cabling.

    At a mid-northern latitude, a 1000 V string using 72-cell modules
    would have 20 modules in series. A 600 V string using the same
    modules would only have 12 modules in series. 1000 V systems will,
    therefore, require fewer combiner boxes to provide the same power
    as a 600 V system.

    This means fewer strings, fuses, disconnects, and
    combiner boxes. This balance of system (BoS) savings can yield a
    reduction of up to 40% in the cost of the DC BoS.

    1000 V-rated wire can cost more than that rated for 600 V, but more
    energy is carried at 1000 V. Increasing the voltage of a conductor
    requires more insulation, but the cost of insulation is much less than
    the cost of copper or aluminum carrying the current.

    1000 V PV systems are standard in Europe because they are
    supported by the European International Electrotechnical
    Commission (IEC). Many manufacturers are now getting dual ratings
    on their products to comply with both IEC and UL standards for
    1000 V, which can also be used, of course, in 600 V applications.
    For these reasons, the availability of products rated for operation
    at 1000 V has continued to grow, and the cost of these products
    has dropped to be more competitive with those rated for 600 V.

    The trend is clear: More designers are choosing 1000 V systems
    owing to the cost savings and other benefits that result from
    having fewer circuits, smaller wire, less labor, and lower losses.
    The applicable codes and standards are also changing to better
    accommodate this growing demand for 1000 V systems.

  23. Tomi Engdahl says:

    380V DC Power: Shaping the Future of Data Center Energy Efficiency

    Deploying DC power distribution in the data center instead of using the traditional AC design is one way to reduce power loss, eliminate unnecessary conversions and, ultimately, lower energy costs.

  24. Tomi Engdahl says:

    DC distribution is not just for the giants

    Facebook and Google are trying rack level DC power. It could be your turn next

    But there is one area where the playing field seems to be level, for both hyperscale and commercial facilities. They are all reliant on electrical power. Further, all facilities receive power in the form of alternating current (AC), and deliver it to their IT equipment in the form of direct current (DC).

    Every data center from the smallest to the largest, is keen to get the best out of its electrical supply, as even minor improvements in efficiency can provide huge savings. For this reason, one issue has come up regularly for the past few years: where is the boundary between AC and DC?

    John Meinecke, president and CEO of EDCS Power explains: “We take a 208V AC drop just like any normal AC-PDU (Power Distribution Unit), and convert the 208V AC to 12V DC at a power shelf in the rack.”

    To make use of the 12V DC at the rack, EDCS Power modifies 1U servers incorporating a 12V DC power supply and backup battery within the 1U confines.

    If the rack contains networking gear (normally 12 VDC ready), Meinecke says the power draw is low enough that EDCS Power includes a single battery backup for all the networking devices in the rack. Meinecke points out that this approach is similar to what Facebook and Microsoft are using in their data centers, and the companies are seeing much-improved efficiencies.

    “EDCS power systems are approaching an efficiency of 92 percent.”

    Enough of an advantage to switch?

    Not everyone is convinced that DC power is the answer. It seems that the return on investment (ROI) is not all that some data center operators want them to be. However, the big guys — Facebook and Microsoft, for example — are convinced, and that has to mean something. And there are smaller outfits giving it a good look.

  25. Tomi Engdahl says:

    The Shift to 380V DC in the Data Center

    Aligning Around the 380V DC Standard

    Support for the 380V DC standard is increasing worldwide, with Japan initially leading the way. DC is already the resident power used by data center subsystems, primarily at the microchip and board level, and virtually all active and passive semiconductor integrated circuits operate from low-level DC voltages. High-speed processors, memory and storage devices, and I/O circuitry all consume DC power, which is readily regulated and distributed in x86-based servers. Lower and lower DC voltages are also being used to power microcircuits, and data processing performance has become increasingly efficient with every successive generation of processors.

    In terms of adopting 380V DC power in the data center, the servers and data center appliances themselves are no longer the conversion break point. Equipment manufacturers are developing technology products that support the 380V DC standard.

    However, for 380V DC to become the mainstream power standard in data centers, three forces must be in alignment:

    Data center managers must embrace the concept, and be willing to make the investment to upgrade their infrastructures;
    Equipment manufacturers must provide the necessary equipment, such as rack power distribution products and servers, to support high-voltage DC power, and;
    Regional utility providers must offer incentives compelling enough for data centers to make the switch.

    There is no question that 380V DC power is gaining traction among managers of greenfield data centers. For example, ABB’s investment earlier this year in Validus DC Systems, a leading provider of DC power infrastructure equipment for data centers, signals that global power company’s intent to drive the market by expanding the availability of DC power solutions for new data centers. But even managers of existing data centers are finding it increasingly hard to resist the considerable benefits of converting to a 380V DC infrastructure.

  26. Tomi Engdahl says:

    Direct current (DC) for data centers

    Direct current power distribution systems are being developed as an alternative to traditional alternating current options. DC system architecture is simpler than that of AC, requiring less space, equipment, installation and maintenance. In addition it can provide improvements in data center reliability and efficiency.

    Using DC in Data Centers

    Technological advances in circuit protection are making DC data centers a reality. In this video, learn how you can make your data center more efficient and reduce the size of equipment by up to 30%, significantly reducing both capital and operating expenses. If you would like to learn more about the solutions that are available and choosing the right protective device for your application

  27. Tomi Engdahl says:

    72V Hybrid DC/DC Reduces Intermediate Bus Converter Size by up to 50%

    Non-isolated IBCs are being designed into many new applications, which significantly reduce the solution size and cost while also increasing the operating efficiency and providing design flexibility.

    Most intermediate bus converters (IBCs) provide isolation from input to output with the use of a bulky power transformer. They also normally require an inductor for output filtering. This type of converter is commonly used in datacom, telecom, and medical distributed power architectures. These IBCs are available from a wide variety of suppliers and are typically housed in an industry standard 1/16, 1/8, and ¼ brick footprints. A typical IBC has a nominal input voltage of 48V or 54V and produces a lower intermediate voltage between 5V to 12V with output power level from several hundred watts to several kilowatts. The intermediate bus voltage is used as the input to point-of-load regulators that will power FPGAs, microprocessors, ASICs, I/O, and other low-voltage downstream devices.

    However, in many new applications, called “48V Direct,” isolation is not necessary in the IBC since the upstream 48V or 54V input is already isolated from the hazardous AC mains. In many applications, a hot-swap front-end device is required to use a non-isolated IBC. As a result, non-isolated IBCs are being designed into many new applications, which significantly reduce the solution size and cost while also increasing the operating efficiency and providing design flexibility.

    Now that non-isolated conversion is allowed in some distributed power architectures, one could consider using the single-stage buck converter for this application. It would need to operate over an input voltage range from 36V to 72V and produce a 5V to 12V output voltage. The LTC3891 from Analog Devices can be used for this approach and provides an efficiency of about 97% when operating at a relatively low 150kHz switching frequency. When operating the LTC3891 at higher frequencies results in a lower efficiency due to the MOSFET switching losses that occur with the relative high 48V input voltage.

    A New Approach

    An innovative approach combines a switched capacitor converter with a synchronous buck. The switched capacitor circuit reduces the input voltage by a factor of two and then feeds into the synchronous buck. This technique of reducing the input voltage in half and then bucking down to the desired output voltage results in a higher efficiency or a much smaller solution size by operating at a much higher switching frequency.

    The LTC7821 merges a switched capacitor circuit with a synchronous step-down converter, enabling up to a 50% reduction in DC/DC converter solution size compared to traditional buck-converter alternatives. This improvement is enabled by a 3X higher switching frequency without compromising efficiency.

    The LTC7821 operates over a 10V to 72V (80V abs max) input voltage range and can produce output currents in multiple 10s of amps, depending on the choice of external components. External MOSFETs switch at a fixed frequency, programmable from 200kHz to 1.5MHz. In a typical 48V to 12V/20A application, an efficiency of 97% is attainable with the LTC7821 switching at 500kHz.

    Capacitor Pre-Balancing

    A switched capacitor converter usually has a very high inrush current when the input voltage is applied or when the converter is enabled, resulting in the possibility of supply damage. The LTC7821 has a proprietary scheme to pre-charge all switching capacitors before the converter PWM signal is enabled.

    Once the capacitor balancing phase is completed, normal operation begins.


    The combination of a switched capacitor circuit to halve the input voltage followed by a synchronous step-down converter (hybrid converter) provides up to a 50% reduction in DC/DC converter solution size compared to traditional buck-converter alternatives. This improvement is enabled by a 3X higher switching frequency without compromising efficiency. Alternatively, the converter can operate with 3% higher efficiency in a footprint comparable to existing solutions.

  28. Tomi Engdahl says:

    Anyside™ Isolated Switch Controller with I²C Command and Telemetry

    Hot swap integrated circuits are used to limit inrush current in either low-side or high-side DC applications. The application operating voltage is generally limited by the integrated circuits operating voltage, typically less than 100 volts. The LTM9100 allows the control of inrush current in both low-side and high-side applications operating up to 1000 VDC.

  29. Tomi Engdahl says:

    LTM9100 – Anyside™ High Voltage Isolated Switch Controller with I²C Command and Telemetry

    he LTM®9100 μModule® (micromodule) controller is a complete, galvanically isolated switch controller with I2C interface, for use as a load switch or hot swap controller. The load is soft started and controlled by an external N-channel MOSFET switch. Overcurrent protection minimizes MOSFET stress during start-up, input step and output short-circuit conditions. Owing to the isolated, floating character of the switch, it is easily configured for use in high side, low side and floating applications.

    A single 5V supply powers both sides of the switch controller through an integrated, isolated DC/DC converter. A separate logic supply input allows easy interfacing with logic levels from 3V to 5.5V, independent of the main supply. Isolated measurements of load current and two additional voltage inputs are made by a 10-bit ADC, and accessed via the I2C interface.

    The logic and I2C interface is separated from the switch controller by a 5kVRMS isolation barrier, making the LTM9100 ideal for systems where the switch operates on buses up to 1000VDC,

  30. Tomi Engdahl says:

    54V DC Battery Short Circuit

    Old BT health and safety video regarding sealed lead acid batteries.

  31. Tomi Engdahl says:

    UL launches data center reliability, safety certification program

    New program to certify reliability of data centers now available. Program from UL and ESD Consulting helps mitigate risk for critical business infrastructure through new UL 3223 Standard.

    Mass adoption of cloud computing has created a significant new risk: the potential to affect large numbers of companies and individuals in the event of a failure of a cloud services provider’s data center, infrastructure or network. To combat this risk, UL approached ESD to collaborate on the development of an international UL Data Center Certification Program. ESD is widely regarded as a thought leader by influencers in the data center engineering, colocation and hyper-scale data center industries, and the company’s expertise enables the engineering services portion of the UL Data Center Certification Program.

    The UL Data Center Certification Program helps mitigate risk for data center owners and operators by providing a set of criteria to increase end-user transparency, provider accountability, and proper data center documentation. The program addresses the continued reliability and safety of data centers by evaluating key components of critical infrastructure. The integration of multiple disciplines creates a comprehensive service for data center owners and operators, who will benefit from the combined expertise of professionals in the technology, engineering, fire and life safety, security, commissioning, and eco-energy areas of focus. Other benefits could include: reduced insurance premiums, reduced construction costs, tax exempt status on personal property and equipment tax and a high-level of marketing brand.

    UL launches data center reliability and safety certification program

    The organization wants to certify entire facilities, and not just their components

  32. Tomi Engdahl says:

    Teardown: The power inverter – from sunlight to power grid–from-sunlight-to-power-grid

    This teardown article will delve into the architectural design and components of a solar inverter card starting from the Solar panel DC inputs and working our way through the DC to AC conversion process to the AC output that is sent out to the power grid.

    In the process we will look at the major elements and component choices that were made in the design of the SMA “Sunny Boy” series of Solar inverters, from the EMI suppression capacitors from Vishay to the TMS320F2812 DSP by Texas Instruments, with a special emphasis on isolation and protection, through the smart use of optically isolated MOSFET gate drivers such as the HCPL-316J and HCPL-J312 from Avago.

    The inverter’s main function is to convert variable-voltage DC from sunlight on the PV panels or battery storage to a specific AC voltage and frequency for use by appliances and feedback to the grid. The AC output varies by region, of course, with 60-Hz 115 VAC used in North America and 50-Hz 230 VAC in much of Europe.

    Enter SMA Solar Technology AG, headquartered in Germany with the “Sunny Boy” series of solar inverters. The inverter board we are looking at in Figure 2 is used in the Sunny Boy 3000TL, 4000TL and 5000TL transformer-less versions rated at 3kW, 4kW and 4.6kW AC output power systems respectively (@230v, 50 Hz).

    Also on the DC input side, two relays are used to monitor insulation resistance in accordance with IEC 61557-8 in pure IT AC systems. See Figure 2 upper left quadrant.

    Measured are insulation resistances between system lines and system earth. When falling below the adjustable threshold values, the output relays switch into the fault state.

    With these relays, a superimposed DC measuring signal is used for measurement. From the superimposed DC measuring voltage and its resultant current the value of the insulation resistance of the system to be measured is calculated.

  33. Tomi Engdahl says:

    AC versus DC load breaking comparison with a knife switch

    Difference in electrical contact by alternating current to direct current.
    The four resistors are wire chrome 650×4 = 2600 watts.
    Note: The tension of 220V rectified is about 311V, but under load it drops to approximately 213-209 volts.

  34. Tomi Engdahl says:

    Power Stamp Alliance to develop standards for cloud data centers

    Now, in 2018, the Power Stamp Alliance emerges with the goal of facilitating market competition by developing product standards based upon STMicroelectronics ICs, especially for the fast-ramping growth of cloud data centers. The alliance’s first product offering is based upon parallel resonant/non-resonant multiphase power converters with current doubler or otherwise known as 48V direct conversion.

    This alliance is different than the POLA/DOSA one. It ensures multiple sourcing of on-board isolated and non-isolated DC/DC converters. How they will put this into practice is to share select information which will ensure form, fit, and function (i.e. a standard form factor and mechanical, feature set, and functional compatibility).

    Although the size of the power solutions must be the same from all members, they will be able to compete with their individual design architectures.

  35. Tomi Engdahl says:

    10 data center power market leaders listed

    According to the analyst, the top 10 leading players in the data center power market include:

    · Emerson Network Power

    · Raritan, Inc.

    · Rittal GmbH & Co. KG

    · Schneider Electric SE

    · Cummins Power Generation

    · Delta Power Solutions

    · Eaton Corporation Plc.

    · General Electric

    · CyberPower Systems

    · ABB Ltd. Among others

  36. Tomi Engdahl says:

    Software/Hardware System Boosts Data-Center Power Availability

    To bolster support for power uncertainty, failure scenarios, and lack of precise operational control in data centers, a SW/HW solution reduces the amount of required reserve power.

    Intel’s Rack Scale Design (RSD) supports comprehensive management for all resources in the data center, enabling a true software-defined infrastructure. RSD data-center architecture separates compute, storage, and network resources into groups of components, called pools, that can be efficiently assembled or “composed” on demand to create a precise hardware configuration to match a specific software workload requirement.

    Because resources are separated and abstracted, each type of resource can be expanded, replaced, or upgraded on its own refresh cycle.

    In most data centers, overprovisioning is a response to several factors:

    Power safety margins required to compensate for the inability to accurately predict power load demands, including peak and seasonal variations.
    Unpredictability, exacerbated by very-high-performance systems, which may drive increasingly large variations in power consumption during peak performance periods.
    Power equipment that’s typically only available (or economical) in large increments.

    Software-Defined Power

    Software-Defined Power (SDP), combined with associated hardware, manages provisioning by employing Intelligent Control of Energy (ICE). This involves monitoring power consumed by IT loads at sub-second intervals using strategically located power sensors. This includes ICE hardware that’s been optimized to provide sensor data.

    Acting as a unit, the ICE hardware makes it possible to exceed, for short periods, the rack or branch circuit breaker capacity, providing peak assurance beyond breaker limits without tripping any breaker. This group policy control is referred to as ICE Rackshare and acts as an application running on the controller.

    The ICE system includes the ICE Block and ICE Switch, a power monitoring and switching system that delivers dynamic redundancy capabilities. It allows unutilized power to be provisioned to additional servers, then reallocated to critical equipment if an outage occurs.

    ICE employs an optimization algorithm that dynamically controls the mix of utility power and local battery power consumed at different points in the data-center topology.

  37. Tomi Engdahl says:

    Power Stamps Poised to Boost Data-Center Productivity

    Next-generation data centers will employ processors, memory, and supporting circuits that require more power for its servers. Enter the Power Stamp Alliance’s proposed higher-power modules.

    Power Stamps are 48-V direct dc-dc conversion-to-POL modules that can provide the higher power density required for future data centers. The new Power Stamp Alliance (PSA) specifies a standard Power Stamp footprint and functions that deliver multiple-sourced, standard modular board-mounted solutions for data centers (Fig. 1). The Founding Members of the Power Stamp Alliance are Artesyn Embedded Technologies, Bel Power Solutions, Flex, and STMicroelectronics.

    These Power Stamps primarily target high-performance computers and servers being used in large data centers, many of which follow the principles of the Open Compute Project (OCP). OCP’s mission is to design and enable the delivery of the most efficient server, storage, and data-center hardware designs for scalable computing.

    The first processor architectures addressed by the Power Stamp Alliance include the Intel VR13 Skylake CPUs, Intel VR13-HC Ice Lake CPUs, DDR4 memories, IBM POWER9 (P9) processors, and high-current ASIC and/or FPGA chipsets supporting the SVI or AVS protocols.

    Serial VID Interface (SVI) is a two-wire (clock and data) bus that connects a single master (processor) to one or more slaves (voltage regulators). Adaptive voltage scaling (AVS) is a closed-loop, dynamic-power-minimization technique that reduces power based on the actual operating conditions of the chip; i.e., the power consumption is continuously adjusted during the run time of the chip.

    This Alliance will mean there will be no single source for modules that combine DOSA (Distributed power Open Standards Alliance) and POLA (Point of Load Alliance) standards. The PSA is similar to both DOSA and POLA. DOSA products share common mechanical pinouts and footprints, and POLA products share common silicon.

  38. Tomi Engdahl says:

    FUJITSU Component Power Distribution Units
    10A-430V DC Socket-outlet and Plug for DC Power Distribution Systems

    DC power distribution systems are gaining
    public interest as a solution to save
    energy. Fujitsu Component offers plugs and
    socket-outlets with various safety functions
    for 400V class DC power distribution systems.
    (Patent pending: patent publicat
    ion number WO2011/102516 A1 etc.)


    Each socket-outlet has magnetic arc-extinguishing module

    Mechanical switch has multiple safety designs:

    Mechanical switch of the socket-o
    utlet works when plug is inserted

    Power flow starts when built-in switch activated

    Plug can not be withdrawn while power is ON

    First make, last break

    approved (File No. E354163)

    Compatible with plugs conforming to IEC TS62735-1

  39. Tomi Engdahl says:

    Direct current (DC)
    plugs and socket
    outlets for information and communication
    technology (ICT) equipment installed in data centres and telecom central
    – Part 2: Plug and socket-outlet system for 5,2 kW

    part of
    6 2735, which is a technical specification
    , applies to plugs and fixed socket
    outlets for class I equipment with two active contacts plus an earthing contact, a rated power
    of 5,2 kW and a rated voltage range from 294 V to 400 V d.c. They are intended to power d.c.
    information and communication technology equipment only.
    The 2,6 kW system complying
    with P
    art 1
    is safely compatible with the system complying
    this part as it is possible to
    the 2,6 kW plug in the 5,2 kW socket
    -outlet but it is not
    possible to
    5,2 kW plug into the 2,6 kW socket

    Accessories shall have a rated power of 5,2 kW at any voltage within the rated voltage range
    of 294 V to 400 V

    From 1,5
    up to
    from16 AWG up to
    14 AWG

  40. Tomi Engdahl says:

    12V vs. 48V: The Rack Power Architecture Efficiency Calculator Illustrates Energy Savings of OCP-style PSUs

    Organizations like Open Compute Project (OCP) and Open19 have created open standards for alternatives to conventional server designs. These architectures redefine how power is converted and distributed within an IT rack. Centralized rack-level power supply units (PSU) replace internal server power supplies. 12V has been the standard output for power supplies, but recently we’re seeing 48V, with claims of much higher overall efficiency.

    Why 48V? It comes down to the quest for improved compute performance, which requires increasing the power to the chip, and at 12V, this isn’t practical. I

    With the default settings in our tool, the efficiency of the 12V rack level PSU architecture is 7 percentage points better than conventional servers (or 33% reduction in losses), and 48V is just over 1 percentage point better than 12V rack-level (7% reduction in losses).

  41. Tomi Engdahl says:


    Our power connectors and cable assemblies for Open Compute Project (OCP) are designed to meet
    OCP power distribution architecture standards. These products provide a simple yet customizable
    design that enables a standardized platform capable of efficiently distributing up to 500A of power
    per UL and CSA criteria, while offering improved electrical performance.
    • Simple plug-and-play solutions
    • Support 12V up to 48V

  42. Tomi Engdahl says:

    Stepping Down From the 380V DC Line

    There is little doubt about the many advantages of the emerging 380V (400V peak) DC power standard that is expected to become commonplace in datacenters and in other telecom and datacom networking facilities over the next few years. Rather than the traditional AC supply, or the 48V DC supply line used widely in the telecom industry, there are many driving factors to move to this new standard including improved system efficiency, reduced cooling and air-conditioning, lower energy consumption, higher power and equipment densities, and increased reliability and availability.

    Standards for 380V DC power systems have been developed by the European Telecommunications Standards Institute (ETSI), including EN 300 132-3-1, which defines the bus voltage between 260 and 400V, and ETSI EN 301 605, which defines earthing and bonding arrangements, as well as contributions from the International Telecommunication Union (ITU-T), the International Electrotechnical Commission (IEC), and the Emerge Alliance.

    The standards define commercial building power system infrastructures that provide power distribution for a telecom and/or datacenter operation within the building. The infrastructure will include a voltage bus and connectors to distribute power to telecom equipment, computers, servers, storage devices, and networking devices, in addition to backup batteries, lighting, fans, and cooling equipment.

    Many suggest the most economical way is to convert the 380/400V DC down to 48V DC, because of the large quantity, or arguably critical mass

    But it could also be that for some equipment used in specific applications, operators may decide to step down to a much lower voltage — for 19-in. rack systems, for example. Many are saying these rack systems should be powered by the 380V line stepped down to 12V (also a commonly used intermediate bus voltage used in telecom).

    Test-systems based on 380V DC have been running for many years now in applications across the globe.

  43. Tomi Engdahl says:

    Eltek launches 400VDC power feed solution for central office powering of remote wireline or wireless broadband systems

    New DC power concept ideal for DSLAMS, small cells; reduces cost and complexity of powering remote equipment cabinets

    Using the HVDC Power Feed solution, mains DC power is converted to 400VDC – a voltage level that can be transmitted long distances with very low losses – at the CO. A second voltage conversion takes place at the point of use, where Eltek’s DC/DC converters transform the 380VDC /400VDC back down to 54VDC /48VDC.

    “Eltek’s remote HVDC Power Feed solution is ideal for wireline and wireless broadband sites that don’t have access to local grid power or in situations where the operator wants to lower power cost and complexity dramatically,” said Bjorn Havard Stokke, Product Manager for Eltek’s portfolio of remote power solutions.

    The Eltek 380V Remote Power solution

    The solution begins with the existing 48VDC power system and battery in the central site. From there, the 48VDC power is converted to 380VDC through Flatpack2 HE DC/DC converters. The 380VDC power then passes through a distribution box providing the necessary protection and safety functions, before being distributed to the load in the remote location.

    At the remote site, another set of Eltek DC/DC converters brings the voltage down to 48VDC for the telecom equipment.

    Depending on the power consumption in each site, and number of remote sites connected to each 380VDC cable, the distance from the central site to the remote site can be up to 5km.

  44. Tomi Engdahl says:

    Powering Norway’s largest infrastructure project
    The longest railway tunnel in the Nordic countries relies on Rectiverter systems to power critical applications

    Rectiverter breakthrough

    Spirits were high and the occasion called for spontaneous celebration in the offices of the Nordic Sales department when the message came through: Bane NOR had decided to go for Eltek’s Rectiverter-based power concept in the Follo Line project. Bane NOR´s main priorities are safety, reliability and performance.

    The world’s first bidirectional, power conversion module
    The Rectiverter is a new concept in power conversion. It combines the functionality of a rectifier, an inverter and a static transfer switch in one bidirectional power module. This opens up for a new power flow architecture and a new way of designing power systems, meeting the needs for both AC and DC output, in one modular system.

    Eltek Rectiverter Concept Animation

  45. Tomi Engdahl says:

    The higher the voltage, the greater is the capacity to transfer power more efficiently and with lower losses. In other words, using 380Vdc instead of 220Vac or 48Vdc, one can transfer significantly more power through the same cable, or the same amount of power through significantly less cable, and with lower end-to-end losses.


  46. Tomi Engdahl says:

    380 Vdc – opening up the route to expansion

    A few short years ago, planners could assume a power level of 750W to 1,250W per rack for the deployment of state-of-the-art servers. Today’s planners project power levels as high as 15kW or more for the ICT equipment that will be delivered in the coming years.
    This is a challenge. Most telecom service providers’ networks are dominated by traditional 48 Vdc power plants that feed the 48 Vdc telecom equipment. Many of the plants are already operating near capacity. Upgrading, replacing or supplementing the existing plants with new 48 Vdc power equipment is, however, not a viable solution.

    Going from 48 Vdc to 380 Vdc

    One of Eltek’s customers, a North American service provider, has taken the step from 48 Vdc to 380 Vdc power transfer, assisted by Robert and his expert colleagues from Eltek.

    – The company in question had a powering architecture with 48 Vdc rectifier plant and batteries in a power room. The 48 Vdc power was distributed to the loads through battery distributing circuit breaker boards (BDCBBs), located in the ICT equipment area. – It was the large cable bundles between the rectifier system and the BDCBBs that created the congestion, Robert explains.

    The customer was preparing for the higher voltage DC future, and had decided to install 380 Vdc rectifiers and batteries at the plant. However, since the existing loads and most of the new loads will operate from 48 Vdc for some time, phase 1 of the strategy was to install a high-efficiency 380 Vdc/48 Vdc converter system in the ICT equipment area.

    Since the plan was to reuse the power transfer cables, it was beneficial to install the 380 Vdc rectifier bays near the 48 Vdc plant. However, the batteries can be located wherever there is space, which can even be in outdoor containers.

    Safe and sound solution

    The technical assessment of the system included lab testing which established that a 380 Vdc system is safe, both in terms of personnel safety and reliability.


Leave a Comment

Your email address will not be published. Required fields are marked *