Power failures, staged power blackouts (like those that Californians now experience) and unexpected disturbances of the utility grid are as painful today as they were decades ago. Diminishing tolerance for power problems and the proliferation of mission-critical applications has resulted in the explosive use of uninterruptible power supply (UPS) systems in nearly every facet of today's business environment. Several different UPS topologies and technologies are available today, but the basic premise remains the same: Maintain continuous power to critical loads subsequent to a power failure while simultaneously isolating them from detrimental power disturbances, i.e. power quality. However, installing a UPS without considering electrical requirements and performance expectations can leave the user with a false sense of security.

One consideration in the UPS selection process is the expectation of availability — that is, how much downtime one is willing to accept. Availability is generally quantified in what the industry has come to refer to as nines. For example, three nines; implies that power to the critical load will be available 99.9% of the time. On the surface, this seems reasonable, but when quantified in terms of a year's time, the 525 minutes of downtime (no power to the load) may be somewhat unacceptable. Likewise, five nines translates to five minutes of annual downtime, which, although not zero, is an improvement of 10 orders of magnitude over the three nines scenario. More nines means more cost. Ultimately, the goal is to reconcile cost with availability as a function of expectation. The amount of downtime one is willing to accept will determine the cost, complexity and overall nature of the UPS system design.

Another consideration in the power protection selection process is the UPS unit itself. Today's UPS, although identical in scope to yesterday's products, have a lot more to offer in terms of electrical performance and system management (connectivity). Selecting the right options for any one application may mean the difference between a basic installation and an advanced solution that ensures maximum availability of power to the critical load. For example, parameters such as input power factor, efficiency and management software are important considerations in the UPS selection process.

Input power factor is a measure of the current distortion the UPS unit reflects back to the source. Low input power factor implies high levels of current distortion possibly causing the backup generator to malfunction, and increases the size and cost of input feeder breakers, input cables and input transformers. Thus, selecting a UPS having a high input power factor (greater than 0.95) and reflected current distortion levels within acceptable limits (maximum 5%) increases availability by ensuring compatibility between input distribution, generator equipment and the UPS module itself. Every UPS manufacturer has its own unique design to obtain this level of performance. Some offer the components to achieve these levels as standard while others offer them as optional. Choosing to have the option installed makes good electrical sense.

Another important consideration in the UPS selection process is efficiency. Efficiency is a measure of the amount of power needed to power a piece of equipment versus what it outputs. For example, an 80kW UPS having an efficiency of 93% will draw 86kW from the utility under normal operating conditions. Selecting a UPS with high efficiency will result in lower utility costs and reduced cooling costs.

Interestingly enough, the same UPS purchased to provide continuous power to a critical load also burdens the utility service, since inefficiencies result in more power being drawn from the source. Thus, selecting a UPS with high efficiency is also socially beneficial since it helps minimize power shortage problems such as those currently felt all across California.

Connectivity is another important aspect of the UPS selection process, but sometimes it goes unnoticed. Basically, connectivity is the software and the associated hardware hookups that permit the UPS system to be monitored across a LAN (Local Area Network), WAN (Wide Area Network) or the Internet. Many different options and accessories are available from the UPS vendors and selecting the right one may seem like a daunting task. However, management of the UPS system is just as important as the management of a group of servers and routers because it serves as the insurance policy against loss of power and ultimately, loss of revenue, loss of productivity and possible hardware damage of the critical load.

Selecting the right UPS system for the job should be done with availability, expectation and performance considerations in mind. Selecting a UPS strictly on the merits of its functionality can result in costly downtime of mission-critical applications.

Mario Recine is a solutions engineering leader with APC, East Providence, R.I.