By David E. Perkins, Chief Technical Officer, Active Power

Demand for UPS

Memories of the 2003 blackout in the Northeast may be fading for some, but for many, this experience illustrated the vulnerability of the nation’s electric grid to conflicts between the surging demands for power in an era of increased resistance to building new power production and transmission capacity. Power producers will attempt to maintain some margin between available capacity and anticipated demand (typically on the order of 12 percent), but unexpected events such as unseasonably warm weather coinciding with planned generation maintenance can have dramatic consequences on electricity availability [1]. Building new capacity has come under increasing pressure due to concerns about the environmental impact of coal fired plants – safety; spent fuel disposal; decommissioning issues associated with nuclear power; and national stability concerns associated with the use of imported oil. Even proposed wind farms have faced challenges for aesthetic and ecological reasons. Some grid managers are now predicting future shortfalls between planned capacity and estimated demand that may result in reduced margins that can influence grid reliability.

We can expect new plants will eventually be built and total blackouts will remain relatively infrequent; however, rolling outages, voltage sags and frequency distortion remain threats to mission-critical equipment that may shut down during such events. Lightning strikes, aging transmission equipment and a host of other factors will always be issues that cause utility disturbances, resulting in power quality and reliability issues that jeopardize critical facility operations. In many cases, the financial implications of power interruptions can be considerable.

 

Business Activity	Outage Costs $/hour
Cellular Communications	$41,000
Telephone Ticket Sales	$72,000
Airline Credit Reservations	$90,000
Credit Card Operations	$2,580,000
Brokerage Operations	$6,480,000

Source: California Energy Commission [2]

 In the past, users have relied on conventional applications for their backup power systems, typically deploying an uninterruptible power supply (UPS) system that depends on batteries. Battery-based UPS systems take up large amounts of space, are high maintenance, environmentally dated and inefficient. These systems run the risk of leaking, are toxic and must be disposed of in a costly and environmentally sound way. Fortunately, there is an alternative to battery-based UPS systems. Energy-efficient, highly reliable and battery-free UPS systems have been developed using integrated, flywheel technology to protect mission critical applications from voltage sags, surges and power interruptions. These systems perform at efficiency levels of up to 98 percent, which reduces energy consumption as compared to lower efficiencies with a conventional, battery based UPS system. From an environmental perspective, wasting less energy means organizations need to produce less. Therefore, this reduces

the amount of CO 2 emissions released into the environment. The flywheel UPS system delivers consistent, predictable performance over its 20-year life span with absolutely no degradation in service.

State of UPS

UPS and backup generator systems provide customers with a second line of defense against grid power quality and availability limitations. According to the Uptime Institute, a single UPS provides power availability of 99.671 percent (mean down time of 28.8 hours per year) and the use of highly redundant UPS systems provides the highest level of protection with availability of 99.995 percent (mean down time of 26 minutes per year) [3].

Historically, three-phase UPS systems suitable for mission-critical applications were of the on-line, double-conversion topology. Double-conversion UPS works by converting AC voltage at the input to DC voltage compatible with the battery, and then back to AC for the customer load. With each conversion stage there is an energy loss. The electrical efficiency of a double-conversion UPS is typically as low as 81 percent at 25 percent load – a common scenario in highly redundant UPS systems. At full load, efficiency is 95 percent or less. On the basis of 1 MW output, the direct losses from the UPS would range from a low of 53 kW up to 234 kW. These losses are dissipated as heat inside the facility must be removed by the HVAC system which can consume another 50 to 100 percent of the direct UPS losses. On an annualized basis, the electricity consumption wasted in losses and heat rejection would be between 0.7 and 4.1 MWh, costing an organization between $69,000 and $411,000 based on electricity cost of $0.10 per kWh. UPS energy conversion waste also taxes the environment by increasing the carbon dioxide load associated with generation. EPA estimates a CO 2 load of 1515 pounds per MWh delivered, yielding a CO 2 footprint between 524 and 3,113 tons per year for this scenario [4].

Double-conversion UPS developed around the only commercially viable energy storage technology available at the time – namely the lead-acid battery. Flooded cell batteries with sulfuric acid electrolyte were initially used, but required extensive maintenance to manage water depletion as well as terminal and plate corrosion. VRLA (valve regulated lead acid) batteries were introduced to reduce some of the cost and maintenance issues associated with flooded cell batteries. However, this has come at the expense of reduced life expectancy and more frequent replacement, and an increase in demand failures (i.e., failure to protect the load when needed, etc.). Chock full of toxic and hazardous materials, both flooded cell and VRLA batteries present safety and disposal issues that have led UPS purchasers to seek a more environmentally friendly alternative.

The addition of UPS systems to reduce anticipated downtime comes at a price. This includes the upfront cost of the equipment and installation, but also the ongoing costs of ownership which includes energy losses and maintenance. There are also tangible costs to the environment from greenhouse gas (GHG) generation associated with wasted electricity production and intangible costs due to continued proliferation of hazardous substances such as the lead and acid content of the batteries.

Just as old Detroit gave way to efficient and clean technology, so will UPS systems eventually be going green as businesses look for incremental environmental improvement. While change may be uneasy, there is reason for optimism. Integrated flywheel-based UPS systems are highly reliable, smaller, more energy efficient and can pay for itself over time.

How Flywheels Work

The principles of flywheel operation derive from Newton’s first law which states that “Every object in a state of uniform motion tends to remain in that state of motion unless an external force is applied to it.” In the case of a flywheel, this motion is rotary instead of linear. Flywheels have been applied in numerous instances where one wishes to obtain a short burst of output with relatively low input. In industry, punch presses and forging hammers were often equipped with mechanically coupled, low speed flywheels to deliver the peak loads required for these operations.

Today’s flywheel systems for backup power are sophisticated combinations of mechanical and electrical technology. Several manufacturers use high strength, aircraft-quality steel rotors for their proven, long-term life performance that has been demonstrated in many other applications. When operated at moderate rotational speeds, simple and highly reliable ceramic rolling element bearings may be used to support the rotor. Bearings of this type have been successfully used in other long-life, mission-critical applications such as aircraft engines, the International Space Station, and other aerospace gyroscopes. Rolling bearings have time dependent failure rates so replacement intervals are specified to assure maximum reliability. Field data from numerous systems shows the failure rate prior to replacement is less than the most optimistic constant failure rate published by manufacturers of magnetic bearings – an enabling technology for high-speed flywheels.

 High-speed flywheels can also be made from metallic materials, but some are now using carbon reinforced plastics (composites). The low density of the composite materials requires they spin at higher speeds than their metal counterparts in order to achieve significant energy storage. This results in low-weight designs which have made them of interest to aerospace systems and electric vehicle manufacturers where minimum weight requirements can justify the higher cost. NASA has been considering the use of composite flywheels for reduced weight gyroscopes and energy storage, but is still studying long-term life performance because the plastic resins that support the fibers can deform over time under continuous application of high stress and elevated temperature during operation.

Integrated within the flywheel rotor or attached separately via shaft is an electric motor that can also function as a generator. The motor accelerates the flywheel to stand-by operating speed and keeps the motor spinning against windage and other drag losses that would otherwise cause the flywheel to slow down. When a utility outage or disturbance occurs, the motor immediately reverts to generating electric power that can be supplied to the UPS. Various motor-generator topologies are in use today, but the most common are the homopolar inductor alternator, permanent magnet or synchronous reluctance machine. The homopolar inductor alternator is less well known, but is similar to the well known and robust synchronous machines used on a diesel genset with the exception that the exciter windings are located on the stator rather than the rotor to allow higher rotor speeds. All are used in conjunction with a variable speed drive that regulates power flow into and out of the flywheel.   

Traditional UPS vs. Integrated Flywheel UPS

The essential elements of the traditional double-conversion UPS are the input rectifier, DC bus and batteries, output inverter, and static bypass. In older units, the output inverter was incapable of delivering line voltage directly so a step-up transformer was used. Newer “transformer-less” designs employ a boost converter to generate the root mean square (rms) AC voltage from the DC batteries (i.e., 480 Vrms from 480 VDC).

 Recognizing the UPS market demand for a better energy-storage solution led to the development of a flywheel solution. Because of the prevalence of battery based double-conversion UPS systems, these early systems were designed as simple DC battery replacements.

Soon after the successful deployment of battery replacement flywheel systems, the idea of integrating the flywheel directly into the UPS was conceived. The advantages of an integrated flywheel UPS system were recognized and these benefits are captured in U.S. patent 6,657,320. The consolidation of functions such as DC bus capacitance, cooling systems, control circuitry, power supplies and display interfaces yields reduced part count for increased system reliability, reduced power consumption for improved efficiency and a significantly smaller footprint compared to comparable power battery based UPS. In addition, the system architecture has only one significant dissipative component in the primary power path – the line inductor. The utility converter path carries only flywheel charging current during nominal input voltage operation and unity output power factor. As system operation deviates from nominal input voltage or unity output power factor, the utility converter current increases, but remains a small fraction of the total system current affecting the system efficiency by approximately one percent. Finally, the DC link voltage between the flywheel and utility converters is higher than the DC bus voltage in traditional double-conversion UPS. This eliminates the need for an output transformer or boost converter to achieve the desired output voltage. The combination of these factors yields a system efficiency of up to 98 percent. In 2005, Lawrence Berkeley National Lab published a detailed study on UPS efficiency and noted flywheel UPS systems had the highest efficiency of all types tested and that the difference between the highest and lowest efficient UPS could span a difference of nearly 15 percent [5].

 Integrated Flywheel UPS Benefits

Efficient, reliable, and green. These attributes set integrated flywheel UPS systems apart from double-conversion UPS with batteries. Consolidation of function and the high power density of the flywheel provide these features in a package that can occupy as little as one-fourth the area compared to the same power systems using the traditional approach. The high energy efficiency and low maintenance are key contributors to operating cost savings that offer return on investment in as little as three to four years.

References

[1] http://en.wikipedia.org/wiki/2003_North_America_blackout. Last accessed June 16, 2007.

[2] http://www.energy.ca.gov/distgen/markets/end_use.html California Energy Commission Web site. Last accessed May 31, 2007.

[3] Turner, W.P, Seader, J.H., and Brill, K.G., Industry Standard Tier Classifications Define Site Infrastructure Performance, Uptime Institute, 2005.

[4] Unit conversions, emissions factors, and other reference data, U.S. EPA, 2004.

[5] Ton, M and Fortenbury, B, High Performance Buildings: Data Centers – Uninterruptible Power Supplies (UPS), Lawrence Berkeley National Lab, December 2005.