Back up your backup power - Technology Information

Demanding Internet infrastructure and deregulated electrical power strain reliability.

With many critical facilities requiring a higher quality, reliability and quantity of power than ever before, enterprises are looking to uninterruptible power supply (UPS) solutions to ensure a stable power environment. Even with a UPS system, careful planning of the UPS configuration is required to achieve demanded zero downtime.

Consider carefully the factors that satisfy the essential design elements: maintainability, availability, expandability and reliability. The first element of a configuration that contributes to system reliability is redundancy.

The redundant UPS system is always fully operational, keeping its own battery plant charged and waiting to support the critical load. Key to this design is that the standby UPS must support a 100% step load with a minimal transient voltage. Redundancy can be achieved with many different UPS configurations.

RELIABLE CONFIGURATIONS

With an isolated redundant configuration, the critical load is never fed from utility power, regardless of UPS shutdown or maintenance. In a traditional parallel system, the static-switch cabinet is between the critical load and the redundant UPS modules. The static-switch bypass is eliminated, and each primary UPS module feeds the critical load on an isolated bus. Should the primary module go off-line, power is seamlessly transferred to the redundant system.

This particular configuration is frequently selected based on its reliability, due to each module's operation being independent of all other modules (primary and redundant), thus eliminating any system-level controls, contributing to unsurpassed reliability. The primary and the redundant modules are provided with a bypass that allows complete isolation for maintenance, while preserving conditioned UPS power to the critical load bus.

Distributed redundant configuration is one of the most prevalent critical power-system configurations selected for use in current Internet data centers. Its most attractive feature is the ability to expand the capacity of the system without interrupting or compromising the flow of critical power to other loads. The configuration uses many groups or sectors of separated primary buses, with each primary bus designed to support a specific area of the data center. There is also a single redundant bus that is typically rated at a kVA equivalent to, or larger than, that of the largest primary bus.

[ILLUSTRATION OMITTED]

The output of each primary UPS module is connected to the preferred input of the static transfer switch (STS). The alternate input of the STS is connected to the output of the redundant UPS system. Under normal conditions, the primary bus feeds the critical load. When the STS detects any deviation in power quality, immediately and seamlessly it transfers the load to the redundant bus.

System availability is improved by the STS. Shutdown of the primary UPS or any deviation in power quality will cause an uninterrupted transfer to the redundant system. The availability of a distributed redundant system allows complete isolation of the primary bus for maintenance by seamlessly transferring the load via the STS to the redundant system. Similar to an isolated redundant system, each module operates independently of all other modules (primary and redundant), eliminating any system-level controls and optimizing reliability. Primary buses can be added to expand the power demand without impacting the existing primary bus or the redundant bus, since all are independent. This allows the system to expand on an as-needed basis.

POWER IS PARAMOUNT

Power demand in most Internet data center applications exceeds what can be provided by even the largest single UPS module. Often, UPSs have to be parallelled for capacity. Additional UPS modules are often added on a parallel bank, to allow for module redundancy. In fact, parallel redundant systems, where one or more modules on a parallel bank may be taken off-line, while the remaining modules support the load, are still the most popular redundant configurations.

The most conventional form of paralleling UPSs is by busing the output with a single static bypass switch. Under normal operation, all the UPS outputs are working in parallel to supply the required load. Should the UPSs be required to transfer off-line, the system-level STS seamlessly takes the UPS off-line, allowing the bypass power (utility or from a redundant UPS) to feed the critical load.

The reliability of the parallel configuration is subject to the integrity and redundancy of the paralleling controls, and whether the parallel bus has redundancy (i.e., allowing one of the UPSs to drop out of the service, still having enough power to service the load; and whether the bypass is fed with protected power from another UPS or not). In many cases, a parallel redundant system will form the redundant bus of a distributed redundant system. As well, it will be used as the building block for many other configurations.

In the end, all that really matters is reliability--in terms of downtime, or the lack of downtime. The first way to determine the reliability of a design is by using empirical or demonstrated data available on a specific design. For example, if the UPS manufacturer has a number of identical existing systems operating in the field and their reliability or performances have been well documented, you can then gauge the expected reliability of similar systems that are going to be installed in similar situations.

Many manufacturers state a mean-time between failures (MTBF) in hours, based on the cumulated operating hours of all the operating units since the last failure of any unit. This calculation--based on empirical data and proven real-life operating situations--can be skewed, however, due to a number of reasons. The first one is inaccurate reporting of failures in the field. Besides, this system also favors the manufacturer with the largest base of installations. Users should be cautious when reviewing MTBF data calculated by the demonstrated operation of the equipment. A more traditional method to calculate MTBF or equipment reliability is using standard MTBF calculations as defined by MIL-STD 217. This calculation is based on statistical failures of individual electronic components that comprise the entire system.

Utility power quality and availability may not address the needs of the growing Internet infrastructure. This places the burden of ensuring power quality on the end-user, who is faced with selecting a UPS topology to satisfy reliability and functionality requirements. The distributed redundant configuration proves to have the necessary reliability level; concurrently, it is also among the easiest and safest systems to expand the capacity under normal operating conditions--a key factor for the scalable Internet industry. Other system configurations have a reliability level equivalent to that of the distributed redundant configuration, but do not offer the same easy maintainability or expandability as the latter.

www.mgeups.com

Circle 262 for more information from MGE

Katz is product manager with MGE UPS Systems, Costa Mesa, CA.

COPYRIGHT 2001 Nelson Publishing
COPYRIGHT 2001 Gale Group