The boon to all you network administrators out there. We luv 'em.

Basically, a UPS is just a large computer battery. These large-scale power devices are run off mains power in to the battery, and then have standard computer power outlets to connect devices in to. The idea is, you run your computers, routers, switches etc off the UPS battery constantly, while it gets its recharging supply from the mains.

The ratings of these batteries are quite large and varied - from small-scale 200VA to the mother of all power supplies, 7500VA. A 200VA UPS will run a small computer for an hour to a couple of hours, based on demand, if power was totally removed from the UPS. The big'uns can run a rack of computer networking equipment for a good 5 hours in total darkness. It always pays to have a torch available for emergency network works.

Lately, the modern UPS has also had remote shutdown facilities included in them. This is usually in the form of a serial cable (or networking variants) and some software installed on the device that you are powering. If there is a blackout, and the batteries are running low, the UPS instructs the equipment to shut down safely. A handy little feature.

Unfortunately, no UPS is smart or mechanically gifted to turn the computers back on when power is restored. *Sigh*

VAs (volt-amperes) are one way to measure the capacity of an uninterruptable power supply (UPS). The VA is the number of volts the UPS can support multiplied by the number of amperes it can support. Yep, you got it - VAs are watts. After my great search for the definition, they turn out to be watts.

When you are choosing a UPS, you need to consider the VA rating and the run-time (and whether it's at half load or full load). For example, if one UPS can run for 45 minutes at 300VA, and another one can run for 5 minutes at 600VA, then the first one is better, even though it has a lower rating. But, if you will be running more than 300 W off it, it's probably better to get one with more VAs.

When choosing a UPS, you need to know how much power your computer uses. An easy way to find this with no additional equipment is to find either the wattage or the voltage and amperes used by each component and convert them to VAs (1 watt = 1 V A, and 1 V * 1 A = 1 V A). Of course, you can take shortcuts - if you don't want to do this for everything inside a computer, you can just use the power supply rating to know the maximum it will be using (although you may upgrade your power supply in the future).

If you expect to add devices in the future, you should add a little to your final result (remember that if you took the power supply rating shortcut additions to your computer won't count, unless you plan to overload the power supply). This will give you the final VA rating you are looking for, and then you just need to determine the runtime of the UPSes available where you are buying. You can either determine the real runtime by calculating it from your computer's VA rating and the runtime (taking into consideration the load it's at - some are half-load to make it look better), by finding the runtime for your computer of one UPS and then calculating relative runtimes of the other ones, or just using pure relative runtimes (this one run twice as long under the same load and I can afford it, so I'll take it).

There are, of course, other ratings for a UPS, such as how well it filters the power (rated in dB), but most people will be buying them for the power backup features, so the VA and runtime ratings will be more important to them. Of course, that "25,000$ surge protection insurance" can't hurt :).

Many thanks to for answering my question and more

An UNINTERRUPTABLE POWER SUPPLY (UPS) is a device for filtering power taken in from mains current and storing it. Various devices which do not react well to power failures are then plugged into the UPS. When the mains power goes off, the device provides power from its own storage to make up for the lack. The amount of current and the runtime provided varies from model to model.

There are two systems which fulfill this function: A UPS, or uninterruptable power supply, in which devices on the load side of the system are always running on battery power (or from other energy storage), which is constantly being recharged; And the Standby Power Supply, or SPS, in which devices normally run off of the mains supply, but switch to the backup stored power rapidly enough to not suffer a loss of power. UPS is used as a blanket term to describe all backup power systems.

The advantage of an SPS over a UPS is that the batteries see less use, and therefore are subjected to less heat, and experience longer life. A UPS, on the other hand, provides superior power filtering, and some particularly sensitive devices might not appreciate the switch from mains current to battery power provided by an SPS. SPSes are of course less expensive than true UPSes.

Many, if not most UPS systems provide some level of monitoring, typically via USB (for newer units) or RS-232 serial connection. It may be as simple as alerting the computer that the UPS has switched to backup power, or it might inform the computer how much runtime is left at the current draw. This allows software which will perform tasks such as shutting down delicate or non-essential systems, either to prevent data loss, or increase run time.

The most popular manufacturer of battery backup systems is American Power Conversion, or APC. They make everything from non-power-storing surge suppressors up to site-installed devices providing enough power for entire data centers. Larger devices plug into 220VAC rather than standard 110, and may provide 110, 220, or both.

While UPSes and SPSes generally use Sealed Lead-Acid (or SLA) batteries to store power, companies are now looking to flywheels to store energy. Active Power, Inc. has a system in which the flywheel is in a chamber "filled" with vacuum, dramatically reducing friction. Unfortunately, flywheels and batteries alike have short service lives of 5-10 years maximum. However, the technology of both batteries and flywheels is improving. Companies are also looking into super capacitors, as they have a very high energy density, though they are also extremely expensive.

Another possiblity for the future is to use hydrogen-based fuel cells - One can apply electricity to split air or water into its constituent components, storing the hydrogen, and putting it back into the fuel cell later when power is needed. Such a device would require a great deal of energy to store the hydrogen to begin with, but would result in high power output and long runtimes. You can also burn the hydrogen in a generator, which implies more physical maintenance, but which may actually be a cleaner source of power over time as the result of burning hydrogen is water and heat, while fuel cells are usually made with expensive metals and must undergo a recycling process at the end of their service life.

UPS systems are available in sizes ranging from about 200vA up to whatever you are willing to pay for. They use batteries of all types, and flywheels to store power, and release it when you need it. Some UPSes are expandable, with the ability to add additional cells to increase run time. You can get a small 400vA UPS built into an outlet strip, or get huge modular power systems which take up more room than a refrigerator - Or indeed fill the entire room. In fact, your local phone company probably still has a room full of batteries providing 48VDC to power the POTS phones in your community.

Log in or register to write something here or to contact authors.