HOW TO SPECIFY A UPS SYSTEM

TO MEET YOUR COMPANY'S EVOLVING REQUIREMENTS

Here is an overview of the key technical issues to address when specifying a UPS system

1.Whole Life Cost

Firstly you must decide on the runtime and the level of availability (or up-time) required of the critical system, estimate possible changes in the magnitude and nature of it in the foreseeable future, and look at the 'whole life cost' (WCL) of the uninterruptible power supply.

A UPS system that can be upgraded and reconfigured in the future could reduce the initial capital investment by not oversizing the system installed. It's worth remembering that estimates of revenue flow, which would be saved by having a high availability UPS system, should be offset against its WCL.

WCL includes: Capital cost of the UPS system. Power density footprint - relates to the rental cost of floor space in the data centre or computer room. Power conversion efficiency (a consideration for On-line topologies only). Maintainability. Scalability - if the magnitude of the critical load were to increase substantially, and if the installed UPS system is scalable, it would not need to be replaced. If a large portion of the critical load comprises blade-servers, which are loads with significantly leading power-factors, then a UPS system designed for the job would be more efficient and cheaper to run.

2. Taking an 'Integrated Systems Approach'

Secondly, another key issue to take into consideration is the provision and quality of related services - planning and installation and preventative maintenance.

One of the inescapable and undesirable side-effects of the trend for higher densities of computers and communications equipment, and UPS systems to protect them, in computer rooms (and data centres and server farms) is the generation of heat. It is becoming increasingly important to take an integrated systematic approach to providing solutions to this problem. Some proprietary solutions are based on air to water heat exchangers.

Integration of UPS system monitoring and control software with other facilities management software is another worthwhile consideration.

Considering the life expectancy of UPS systems and their components

The life expectancy of any UPS system is, to a large extent, dependent on its ability to be right-sized to the critical load. Some rackmount modular UPS systems can easily be scaled up, or down to meet changing demands. Others do not inherently have this capability.

UPS systems comprise electronic assemblies, lead-acid batteries, cables and cabinets. With routine maintenance, the electronic assemblies, cables and cabinets will last indefinitely. However, batteries, which store the standby energy, have a useful working life of typically four or eight years (dependant upon 'design life'). In general, life expectancy of lead-acid batteries diminishes with exposure to temperatures of over 20 Celsius, with each 'discharge/recharge cycle', and by being in a state of 'low charge' for prolonged periods. However, intelligent state-of-the-art UPS systems ensure that their batteries are fully charged when the mains AC supply is available, and do not discharge to 'deep discharge' levels when it is not.