Download document () of 20
Send email
Download
You have exceeded the download limit
In recent years, many data centers have adopted a zoned strategy, where the data center is divided into zones each powered by a UPS system. This approach is not only more scalable than a centralized UPS strategy, but a UPS maintenance event or failure only affects a single zone rather than the whole data center. The zone concept has advanced further with distributed architectures, where each rack or set of racks has its own UPS system—and the UPS itself is modular to expand as needed. This approach is ideal for colocation and hosting providers, they can segment customers and scale out only when there is demand.
Redundancy can be established through paralleling or modularity to establish Tier II and higher reliability. In UPS paralleling, two or more UPS systems are electrically and mechanically connected to form a unified system with a single output—either for extra capacity or redundancy. In modularity, a single chassis can be populated with multiple capacity or battery modules to provide a self-contained and compact redundant power system. There are many factors to consider when determining which redundancy approach is best for a given application. As a general rule of thumb, paralleling works well for three phase applications and modularity works well for single phase solutions. In an N+1 redundant configuration, you would have at least one more UPS module than needed to support the load. As a conjoined system, each UPS stands ready to take over the load from another UPS whenever necessary, without disrupting protected loads. Paralleling provides an excellent solution for matching growth while extending the value of existing UPS systems. Certain UPS systems are also capable of multi-tiered levels of redundancy, which is often noted as N+X.
Modular-based solutions are the focus in the single phase 4-20kW power quality space. Modular chassis solutions can be free standing or mountable in standard 19-inch IT racks. Because these units are built around a single frame, the only professional installation required is the initial electrical wiring or rack mounting of the unit. Once these solutions are deployed, capacity and runtime upgrades can nominally be deployed by the end user with no third-party assistance. These compact designs allow for easy upgrading and help minimize installation, cooling, and maintenance costs over the deployment life of the UPS.
Typically, in modular systems, each power module (capacity) contains its own internal static bypass and can support the load of another power module in the event of their failure. It should be noted that this behavior normally forces the UPS to bypass, however it does prevent load drop. Maintenance bypasses are usually installed in the rack with the UPS or occasionally wall mounted nearby if there is no rack space to accommodate. These features make modular UPSs an ideal solution for single phase end- of-row and network closet applications up 20kVA with N+1 redundancy. When coupled with a cyber secure networking card and remote management these solutions offer unrivaled flexibility, monitoring, and upgradeability on any budget since an end user is never paying for more than they need.
Today there are power quality solutions that are compact, rackmount and expandable from 8 to 60 kW in a single 19-inch rack that can also reduce energy and cooling costs. These units reduce complication with deployment and can be installed without an electrician like the Eaton BladeUPS.
Each UPS system can have the same paralleling technology as larger systems and each module contains its own internal battery trays, built-in static switch, maintenance bypass capability and hot swappable components for ease of replacement or upgrade. The intrinsic reliability of the multi-megawatt parallel systems is preserved in smaller capacity systems.
You can now get UPS systems with higher power ratings in a compact design with internal redundancy and embrace the scalable configuration at the 20-400 kW level.
A new trend to distribute critical power at the end of each row reduces upfront capital expenditure (CAPEX) and allows for unforeseen future growth with vertical scalability at the row level.
End-of-row designs reduce single points of failure by placing the critical power next to the critical loads. When it’s time to add more capacity or even replace components, the smaller units allow for quicker deployment, lower incremental costs, and improved availability.
A common concern when putting a uninterruptible power supply of this size in the white space is airflow management and the lack of floor placement options, but UPSs of this size should provide maximum deployment flexibility by providing innovative thermal management options, flexible configurability and full service access from the front.
A modular power infrastructure makes it easier to support scalability in IT equipment. It is crucial, however, to plan for the future. When evaluating components for data center power infrastructure, look for the ability to support modular expansion and redeployment later.
