UPS systems have thoroughly outgrown their origin as simply ‘power
protection’ units. The sophistication, significance and sheer scale
of the equipment and processes they support has led to equally
sophisticated UPS technologies, to the extent that they have moved
from ‘backup’, to the forefront of energy management and
UPS technology advances and system flexibility mean there are now many more opportunities to optimise installations for increased efficiency and availability, or to reduce costs and emissions. Rising energy demand and pricing, along with pressure to shrink carbon footprints, make these achievements all the more desirable.
The proliferation of microprocessor-based equipment in industrial and commercial sectors alike has dramatically increased the numbers and types of electrical load falling into the ‘critical’ category.
For example, the majority of businesses are now heavily reliant on the internet and data processing, for internal operations as well as customer services, and require sizeable IT infrastructures. Expectations of 24/7 availability and the opportunities for round-the-clock global trading and productivity, have in turn driven the need to maximise system uptime, since even minor interruptions to business can cause a significant financial loss.
Uninterrupted mains supply is essential for business critical systems at a time when the National Grid is under pressure to meet growing demand with an aging infrastructure and uncertainty about future energy sources. Independent power protection – and standby generation – are sure to be an increasingly indispensable element of businesses’ energy strategies.
Of the three main types of UPS system, off-line, line-interactive and on-line, implementations of on-line double-conversion UPS systems have proven most effective, especially with critical loads such as computer rooms and data centres, because they offer the most comprehensive protection against mains supply problems and consequently deliver the highest levels of system availability.
Transformer-based on-line UPS, which utilise an internal step-up transformer, were developed in the 1970s and are still in production, particularly at the very top of the output power range. However, with advances in power semiconductor technology, transformerless three-phase UPS systems were introduced in the early 1990s and are now widely adopted, delivering significant weight and space savings and enabling the development of today’s modular, rack-mounted UPS.
Such systems provide high efficiency and power density, and the smallest physical footprint on the market. Compared with traditional free-standing units, they take up only 25 percent of the floor space, and vertically scalable modules mean that additional capacity for redundancy or load upgrades can be easily achieved at a fraction of the cost of an additional stand-alone unit.
There is huge potential to reduce electricity consumption, and to alleviate the burden on stretched cooling systems, by continually matching the capacity of UPS systems to their respective critical loads. Modular, transformerless UPS systems are much more flexible than their traditional counterparts at matching load requirements and delivering optimum efficiency.
Trying to cater for future needs with stand-alone UPS systemscan also lead to over-specification, creating a wasteful gap between installed capacity and the size of the actual critical load. Such inefficiencies mean that companies could be burning excess electricity and creating needless heat emissions, compromising efforts to control costs and reduce their carbon footprint.
Figure x (handbook p230/31) illustrates how the ability to 'right-size' modular UPS systems promotes efficiency. The limited flexibility of a stand-alone UPS would require the initial installed power to exceed anticipated capacity requirements, resulting in an inefficient, over-sized system. However, the flexibility and scalability of modular rack-mounted systems means they can be ‘right-sized’ from the outset by inserting or removing ‘hot-swappable’ modules, enabling power to be cost-effectively added as requirements grow and without any footprint penalty.
Since power problems are the largest single cause of computer downtime, increasing power availability is the most effective way to increase overall systems availability. The single most important issue in increasing power availability is to decrease the mean time to repair (MTTR) of the power protection system.
In figure x (Fig 2 in C28144 FAT article), the availability of a traditional UPS system and an advanced modular UPS system are compared. The UPS system on the left comprises two 120kVA free-standing UPSs in 1+1 parallel-redundant configuration, and the one on the right comprises four 40kVA ‘hot-swappable’ UPS modules in 3+1 parallel-redundant configuration.
Their Mean Time Before Failure (MTBFs) are 600,000 and 400,000 hours, and their MTTRs are 6 hours and 0.5 hours respectively. However, the availability of the free standing solution is 0.99999 (five nines) while the modular solution provides an availability of 0.999999 (six nines). This higher availability increases overall system availability by a factor of 10 compared to free-standing (non-modular) UPS systems.
Parallel UPS systems comprise either centralised parallel architecture (CPA) or decentralised parallel architecture (DPA). While CPA systems offer a cost benefit by sharing common components, the drawback is that this centralised configuration introduces a number of ‘single points of failure’ into the system, which adversely affect its availability.
UPSL pioneered the development of advanced decentralised parallel architecture, paving the way for the high-power PowerWAVE 9000DPA system, now complemented by the new 8000DPA for small to medium applications (10kVA to 120KVA).
As well as valuable efficiency benefits and low total cost of ownership, decentralised UPS configurations offer maximum system availability. Paralleled and independent UPS modules, containing all the hardware and software required for full system operation, eliminate potential single points of failure by cost effectively duplicating all critical components within each module. With a minimum of one module over and above that required by the ‘capacity’ system, the load is supported with inverter power if any one module shuts down, thereby increasing system reliability and giving guaranteed system uptime.
DPA technology undoubtedly comes at a premium, but the lower purchase price of traditional UPSs must be offset by significantly greater operating expenses. Reductions in energy loss costs with modular DPA systems mean the higher initial outlay can be recouped within the first year of operation, with further savings achievable in the longer term – more than £25,000 over five years in the example illustrated by figure x. In addition, the elimination of single points of failure means that modern UPSs also provide enhanced protection against revenue losses and costs caused by system failures.
Understandably during a recession, self-preservation may have momentarily taken priority over saving the environment (business closure being an undesirable means of ‘cutting emissions’). However, taking care of the bottom line now more often involves reliance on the continuous uptime of business critical IT, communication and production facilities. UPS systems, thankfully, can simultaneously support financial and environmental goals without compromise.
Typical media images of factories belching pollution belie the fact that technology intensive businesses, and data centres in particular, are responsible for a significant proportion of greenhouse gas emissions. According to a 2008 McKinsey report, today’s energy intensive data centres emit as much carbon dioxide as the whole of Argentina, with output liable to quadruple by the year 2020.
The Climate Change Act 2008 includes a statutory target to reduce UK greenhouse gas emissions to at least 80 per cent below 1990 levels by 2050. Latest projections suggest that net UK greenhouse gas emissions will be 36% below the 1990 level by 2020, and are on a path to meet the 2050 target.
However, Uptime Institute research suggests that energy consumption in the top third of surveyed data centres grew 20 to 30 percent annually in 2006 and 2007, so there is still a clear onus on businesses to adopt more rigorous energy efficiency strategies.
While poor utilisation of servers and facilities is a major cause of inefficiency, modern UPS technology offers significant opportunities to improve not only energy consumption but also to reduce emissions. Modular on-line double-conversion UPS systems are helping organisations become more energy efficient, qualify for tax incentives and comply with government initiatives to promote energy efficiency.
Business research and consulting firm Frost & Sullivan has highlighted rising energy costs, declining power quality and concerns over carbon emissions, and notes that it is vital for applications consuming high levels of power, such as data centres and industrial applications, to adopt energy-efficient UPS.
In fact, UPS customers are increasingly seeking modular, rack-mountable solutions, not only to maximise system flexibility and availability, but to meet wider strategic aims on energy efficiency and environmental compliance.