With most organisations now wholly dependent on their data, UPS is a vital requirement for every resilient business. A partial or total failure can have catastrophic consequences including financial impact and loss of reputation.
Data protection should form part of a continuity plan which is put in place to protect the business when a situation arises that may result in a loss of power. There are three categories that certain loads fall into depending on what purpose the data serves.
The more central the computer and telecommunications systems are to the product and service, the more they can be classified as critical loads. Critical loads include data centres, hospitals, mission critical operations, airports, and telecommunications companies.
For most organisations, the load can be grouped into classes known as essential and non-essential. Essential loads require a form of back-up but do not require uninterruptible power and can be allowed to fail or ‘ride through’ – this is the time it takes for the alternative back-up system to start up, such as air conditioning, heating and emergency lighting. There are also non-essential loads which are those an organisation can afford to lose when the mains power supply fails for a period of time without a detrimental effect on the business, such as general lighting and some print services.
Depending on the requirement and the power load, businesses can easily buy an ‘off-the-shelf’ UPS which they can install themselves. However, for large systems, the assistance of experts is vital as they can help in selecting the right location for the system and ensure it is installed correctly.
There are three main types of UPS: offline (or passive standby); line interactive; and online or double conversion. Each of these supplies energy from the battery when the grid fails, but under normal conditions they process the power differently.
Off-line/passive standby UPS
Off-line UPS is the simplest and least expensive type of UPS system on the market and is normally limited to the protection of applications such as work stations, terminals, and low power communications devices below 1kVA at 230V AC single-phase.
It is compact, designed for desktop or wall mounting and uses its own output to track voltage and frequency variations in the mains supply power so it is primed and ready to power the load for a short duration (10-15 minutes) from its built-in battery in the event of a power cut.
Most power cuts last less than a few seconds, which means the load is protected by this type of uninterruptible power supply. In the case of an extended power failure, this type of UPS is designed to perform a controlled shutdown of equipment to negate potential damage but it is not designed to provide extended runtime.
Line interactive UPS
A line interactive UPS uses built-in passive electronic regulation devices to stabilise and regulate voltage fluctuations. When mains supply is present, the output frequency of the UPS tracks the input to the mains.
The inverters within line interactive UPS vary in design and output waveform. They can be designed to supply a sine wave or modified sine wave known as a step-wave. Whichever waveform is supplied, the inverter output will typically be regulated to within 5%, i.e. much wider than the output regulation provided by an on-line UPS.
Line interactive UPS have internal battery sets typically sized to provide a five-minute runtime at full load. Longer runtimes require either over-sizing the UPS for the particular load or an inverter rated for longer operation at full load, which may be the case for the larger sized UPS in a particular range. Where this is the case, the UPS will have a rear panel connection for external battery extension packs. However, unless the charging capacity is also increased, the recharge time may exceed the nominal target of 80% within 12 hours.
On-line or double conversion UPS
On-line or double conversion UPS maintains voltage and frequency variations within prescribed limits so that output is independent of fluctuations in supply. On-line UPS are the preferred choice for critical data and voice processing systems because they provide superior electrical performance (whether on mains or battery power), a break-free supply on mains power supply failure (or restore) and an automatic system bypass for safe failure to mains if an overload or fault condition occurs.
There are two approaches available with on-line UPS: transformer-based and transformer-less. Both produce a tightly regulated source of uninterruptible power and primarily differ in the way they generate the DC busbar voltage required by their inverters and therefore their output stages. This leads to differences in their physical size, efficiency, noise output and the levels of input harmonic distortion they generate.
Transformer-based on-line UPS are the most traditional design and are available in sizes from 8kVA to 800kVA. The most common applications for this type of UPS tend to start above 120kVA. This type of UPS has a robust transformer isolated inverter output making it more suited to protect applications on, for example, industrial sites where there is a high degree of electrical noise, spikes, transients and potential high short-circuit currents.
The transformer-less design is the more modern of the two and is commonly available from 700VA to 120kVA. The design has become the standard for most IT environments below 120kVA. Most of its advantages derive from the fact that this type of UPS does not have a built-in transformer. It is therefore more compact, with a smaller footprint and generates far less noise and heat, and lower input harmonic distortion levels.
As the two UPS ranges overlap in terms of power rating, selecting the right one for an application is not always easy. The solution is to review both types and their scientific advantages in relation to initial purchase costs, physical footprint, running costs, the actual installation environment and, in particular, the levels of input harmonic distortion they generate.
Having identified the critical, essential and non-essential loads, the choice is down to load category, load size and the level of resilience required and the need for redundancy. UPS systems are grouped by manufacturers into specific applications, which are: IT; network; industrial and enterprise; and rackmount.
IT applications 300VA to 3kVA
IT typically covers smaller applications - information and communications technology (ICT) typically, including home PCs, small office/home office and data and voice networks. UPS for this type of application include off-line, line interactive and on-line designs up to 3kVA.
On-line uninterruptible power supplies in this range are transformerless so as to achieve a small footprint, minimal weight and low noise and heat output. They are usually located in computer rooms, next to protected equipment. However, their reduced size means fewer output sockets, so in highly populated environments a greater number may be required, which will have an impact on capital and operating costs.
Network applications 3 to 80kVA
Network applications include corporate data and voice networks such as those run by ISPs (internet service providers) and telecommunication companies. UPS systems for this type of application are transformer-less, which, again, minimises size and weight.
They act as a centralised power source and are hardwired at both input and output due to the high levels of power required. The UPS may also require connection to a three-phase incomer. The loads themselves are more likely to require three?phase and dedicated power distribution switchgear.
Industrial and enterprise applications 10 to 800 kVA
Industrial or enterprise applications, upwards of 10kVA, are usually for enterprise?wide data and voice networks, industrial processes, security (emergency lighting, fire and security systems) and large hospital applications. Traditionally, transformer-based UPS systems, with either a 6-pulse or 12-pulse rectifier fitted as standard, provided robustness but were less efficient, generated higher heat and noise levels and were contained in larger, heavier systems.
Thanks to continued research and development for this growing sector of the market, modern UPS systems now utilise the very latest IGBT (insulated gate bipolar transition) rectifier technology in conjunction with a high performance transformer. This offers higher efficiencies, low harmonic generation, high input power factor of 0.99, and a much smaller footprint, whilst maintaining an extremely high level of resilience.
Rackmount applications 700VA to 30kVA
Rackmount applications start from around 700VA all the way up to 30kVA. Rack cabinets have become a common installation format for multiple server projects to reduce space and cable runs in computer rooms. Manufacturers of UPS have responded by creating rackmount formats from standard product ranges.
UPS can be installed in racks that are strong enough to take their weight. A typical rack cabinet has a loading of 150?300kg. In this instance, the UPS often sits next to battery packs (part of the power protection installation) and protected loads. Sometimes UPS and batteries will have their own, dedicated rack.
Correctly sizing the UPS load and matching this to the right UPS size and topology is an integral aspect of a power continuity plan. There are, however, additional considerations and in today’s environment, a key decision has to be made as to the type of UPS design to deploy. Businesses still unsure of what UPS requirement they need are advised to seek advice and, dependant on size, have a professional perform a site-visit to make sure what they are getting is right for them.
The future of UPS
In terms of power protection, UPS plays an important role in improving energy efficiency and manufacturers have invested heavily in developing and introducing products with high efficiency ratings, as well as finding ways of ensuring the eco credentials of power protection systems. With more companies looking at greener ways to operate, Riello has turned some of its attention to flywheel UPS technology.
A flywheel is a mechanical device that rotates at incredible speed, 20 000 – 50 000rpm in many cases, and produces kinetic energy, which is then stored. It works by accelerating a rotor using a source of mains electricity and maintaining the energy in the system as kinetic energy. The amount of power stored in a flywheel is proportional to the square of its rotational speed.
Flywheels provide a source of short duration dc power and can be used with large applications such as UPS from around 60kVA upwards. They can be used in place of a battery set to provide the 10-45 second ride-through time required for the automatic starting of a standby generator, or simply reduce the initial discharge of a battery set.
In terms of UPS, it needs to be compact, powerful, reliable and resilient, which is why flywheel UPS technology is attractive because of its compact size in comparison to battery banks, high-energy efficiency, low maintenance, lifetime cost, silent operation and zero emissions. Flywheels can also reduce the need for costly cooling within the UPS vicinity.
Modern flywheels, made of composite materials, are lighter in weight so they do not require the costly and disruptive re-enforcement of flooring at installation. Being lighter means they can also spin faster and thus produce more energy in a shorter time. The capital cost of purchasing a flywheel UPS is higher than a comparatively rated traditional UPS but over its lifetime it can save money due to the fact that it negates the requirement for battery replacement and heavy-duty maintenance.
Flywheel technology has advanced beyond the large, industrial solutions of yesteryear. They are now commercially viable for UPS application and data centres in particular, in which the necessity to reduce energy consumption, address ‘green’ initiatives and reduce total cost of ownership is equal to the need to provide resilient and scalable power protection for critical applications.