Designers and manufacturers find new ways
to work with and beyond traditional battery storage technologies to support UPS systems and bridge power supply prior to generator startup
Let's face it: The grid can be unreliable. Evergreen, Colo.-based Research Reports International estimates that nearly 20% of underground transmission is more than 30 years old, and overhead lines are constantly compromised by extreme temperatures, the weight of ice, and strong winds and storms. Overburdening during peak hours causes periodic voltage swings, frequency changes, and intermittent outages due to brownouts. Even more alarming, the National Electricity Reliability Council predicts that more than 35 states will be operating with capacity margins less than 10% by 2009.
All of these conditions can cause costly downtime for commercial businesses, manufacturing, and industry. Energy Power Research Institute, the Palo Alto, Calif.-based independent, nonprofit center, estimates that U.S. businesses lose up to $30 billion a year from power interruptions. In addition, researchers at the Lawrence Berkeley National Laboratory, Berkeley, Calif., report that electrical power outages and blackouts cost the nation close to $80 billion a year, with the commercial sector accounting for $57 billion of the total, and the industrial sector accounting for almost $20 billion.
It's no wonder the sale of power quality products is on the rise. In fact, Norwalk, Conn.-based Business Communications Company, Inc., predicts that the U.S. market value of major power quality-related products is poised to jump from $3.8 billion in 2000 to $7.1 billion through 2006, an average annual growth rate of 11% — with uninterruptible power supply (UPS) systems comprising the bulk of that revenue. Palo Alto, Calif.-based industry research analyst firm Frost & Sullivan projects that revenue from the sales of UPS systems will grow from $5.29 billion in 2001 to $6.29 billion by 2007.
UPS systems provide a finite source of electrical power to support selected critical loads during a loss of normal power. Their backup times range from seconds to hours, or at least enough time to switch over to an on-site generator in the case of extended outages. Most UPS systems generate independent power with the use of batteries, but recent breakthroughs in energy storage technology and renewable resources may change that in the future. For the last decade, designers and manufacturers have been working to reduce the cost, maintenance, and footprint, as well as increase the energy density, of battery alternatives, such as flywheels, compressed-air systems, and ultracapacitors. Compatible with most energy generation and consumption systems, including renewable energy sources such as wind and solar power, these devices can be used to ensure good power quality, as well as store energy during times of low demand, thus helping reduce power system loads by providing off-grid energy during peak demand.
“The two biggest problems that customers tell us they have with batteries is reliability and cost,” says Joe Pinkerton, chairman and CEO of Austin, Texas-based Active Power. “In 3-phase UPS markets, in essence, you're putting a minimum of 40 car batteries in series — any one of which goes down bringing the whole thing down — so the mean time between failure is really measured in months.”
Reliable battery performance depends heavily on adequate maintenance. Lack of proper maintenance may lead to premature replacement and high lifecycle costs. However, there is some resistance to the leap from batteries to the use of these other devices. Facilities managers and engineers are comfortable with battery-based systems and may have a difficult time justifying the change in power quality products to their management. “No one ever really lost their job because they replaced a room full of batteries again,” says Mike Everett, chief technology officer for San Diego-based Maxwell Technologies. However, given time and further improvements in technology, newer power quality products may one day replace battery-based systems altogether.
Batteries included. Since their introduction in the mid-19th century, batteries have been the most commonly used devices for storing electrical energy. Of all battery types, the most frequently used is lead-acid, made up of plates, lead, and lead oxide submerged in a 35% sulfuric acid and 65% water electrolyte solution, which is the catalyst for the chemical reaction that produces electrons. Similar to lead-acid, flow batteries also use an electrolyte solution, which is stored in external containers — as large as needed and placed where most convenient to the end-user — and circulated through the battery cell stack.
Variations to the lead-acid design are valve-regulated lead-acid (VRLA) batteries, which are sealed and don't require topping off with water, and gel-type lead-acid batteries, filled with a gel instead of liquid, making spilled battery acid a less likely occurrence. Advanced battery technologies include lithium-ion, lithium polymer, nickel metal hydride, and sodium sulfur. These battery types have a much smaller footprint than their lead-acid counterparts, but currently are typically too expensive for large-scale cell applications.
Batteries require frequent maintenance and safety checks, as well as room for their footprint, depending on how much energy needs to be stored. For example, the average telecom application requires multiple redundant strings of 24 cells, but a data center uses batteries in 240-cell strings, so the risk of failure is high because failure of any one battery cell brings the entire string down. Premature failure can also result from heavy cycling loads and fast rates of charge and discharge. Lead-acid batteries also require proper disposal and recycling due to their lead content.
It's these drawbacks that provide the impetus for alternative devices for energy storage. However, batteries are in no danger of becoming obsolete any time soon. “Wherever you need long-term energy, a battery is going to find a home,” Everett says.
It's with this in mind that most designers and manufacturers have engineered their new products to be compatible with the traditional power quality standby or to replace them only on a case-by-case basis.
The “mechanical battery.”. One technology used to either extend the life of batteries or altogether replace them is the modern flywheel. (See sidebar below.) The earliest flywheels, such as the potter's wheel, have been in existence for thousands of years. The devices have also been a part of engine designs for hundreds of years. Since the late 1990s, modern flywheels became a commercially viable device for overriding momentary voltage and frequency changes, as well as bridging energy sources — from grid to backup generator — during extended power outages. Used in parallel with VRLA batteries, flywheels can “harden” battery banks, taking hits from the short-term discharges and step loads, thus extending battery life.
The relatively short backup times have put flywheels in a specific position within the power quality field. They are most successful in applications, such as manufacturing, where customers want a reliable transfer of the critical load to the backup genset. But even in these “bridge” applications, some customers feel more comfortable keeping their batteries. “At the end of the day they still need to report to their management and make sure that they made wise choices,” says Jean-Philippe Poirrier, director of marketing at Chatsworth, Calif.-based Pentadyne.
Poirrier agrees that there are two types of customers for flywheels: end-users who use flywheels to extend the life of their battery systems and those who use the flywheels as a bridge to their genset. “If it's really critical, you are going to have gensets,” he says.
If you have a genset, you only need about 10 seconds for a seamless transition. Because of this, the objective for flywheel R&D isn't necessarily extending the device's storage times. According to Poirrier, customers with gensets are realizing that a full 15 to 30 minutes of bridge time isn't necessary. They are the first group to forge ahead without batteries. The flywheel's current run time is sufficient to bridge the gap between an outage and switching to a genset.
Replacing batteries with a flywheel that has a significantly smaller footprint also makes a lot of sense. “When you get a facility manager that can report to the top management, ‘You know guys, we can rearrange the setup and recover footprint space,’ that rings a bell for people who need expansion but don't have means to go someplace right away,” Poirrier says.
Running hot and cold. There are facilities, such as mid-sized data centers, that don't see the need to invest in an on-site generator for backup power. In the absence of a genset, the typical flywheel storage time isn't adequate to ride out longer power outages. Finding it impossible to build a bigger flywheel or concoct a system out of several flywheels to extend the run-time to 15 minutes — the equivalent of a battery system — designers at Active Power came up with the thermal and compressed-air storage (TACAS) system, combining elements of both the flywheel and compressed-air energy storage. (See sidebar below.) “Customers were demanding it,” says Pinkerton. “We pounded our heads for several years trying to figure out a way to provide them what they wanted.”
The system can compete minute-for-minute with batteries, but without the hazardous materials or environmental issues associated with the acid and lead contained in batteries. A single module can provide up to 15 minutes of backup power at full load or can be configured to provide up to several hours for lighter loads. The cool-air system also has an added advantage for datacom applications.
“We're working on a device that will literally measure the temperature of the room that it's in, and if it senses that things are getting hot, it can create more and more cool air,” says Pinkerton. “Even to the extreme that if it's supporting an 80kW load, it literally can produce enough cold air to absorb 80kW of heat during the discharge.”
Currently, the base product provides 85kW, so it's used mostly in commercial applications. In the future, there may be a 10kW version, which could possibly be used in residential applications or — the more probable scenario — in the telecom industry. But the R&D won't stop there. With the relatively new product — the beta units shipped in the last year — there are many avenues to explore. “You could go up in power,” Pinkerton says. “It might be cost-effective if we develop a 1MW system that could be applied both to backup power and to these utility load leveling applications where people want to store cheap energy at night and deliver it during peak demand times in the afternoon. So it's go up in power, go down in power, and add features.”
Burst power. Supercapacitors, or ultracapacitors, are considered power devices, whereas batteries are energy devices. Power devices are most frequently used in applications that require burst power, or burst energy — strong, short spurts of energy to get you through a step load or bridge power to a generator startup. (See sidebar below.) The devices also can be suitable partners to batteries by taking hits from surges, thus extending the life of the batteries.
“An ultracapacitor module that's built around a 48V telecom central office station can provide up to a minute and a half of backup power while you transition from the grid to the microturbines or gas generators,” says Mawell Technologies' Everett. “It can replace a room full of batteries.”
The most convincing selling points of ultracapacitors are a lifecycle lasting 10 to 15 years and little to no maintenance because they're hermetically sealed; not allowing air or water in or out. Theoretically, ultracapacitors may be charged and discharged millions of times, compared to a mere thousands of times for batteries. In addition, ultracapacitors have a low-temperature operating capability — as low as -40°C, which makes them ideal for harsh environments.
Other than adoption, the biggest challenge for ultracapacitor manufacturers right now is cost. “We're all working hard to get the cost of these devices down to where any manufacturer in any market won't think twice about using them,” Everett says. “Only in the last five years have ultracapacitors started to come down in cost for manufacturers to build them — and down enough in price to find a way into applications.”
The second motivation for ultracapacitor designers and manufacturers is to improve the energy density of the devices. Energy density for ultracapacitors is measured in similar terms as batteries: You want to get higher voltage capability out of the capacitors. Currently, ultracapacitors are limited to 2.7V operation. For a 15V system, designers and engineers have to put six ultracapacitors in series. So the cost of system implementations is also directly proportional to the operating voltage.
“The facility managers need to be attuned to the fact that it's a lifecycle cost asset rather than a short-term cost benefit,” Everett says.
But facilities and plant managers may have deeper-rooted reasons against replacing their batteries. “There are historical reasons why people are reluctant to adopt a new technology,” says Everett. “Lead-acid battery backup systems have been around forever. If they put some ultracapacitors in there, they're taking a risk. So it's really a kind of a risk adversity that's the stumbling block between adoption and the choice not to.”
Sidebar: Building a Better Flywheel
Flywheels are cylinders that spin at high speeds, storing kinetic energy. The faster the flywheel spins, the more energy it retains. Energy can be drawn off as needed by slowing the flywheel. They can be combined with a device that operates either as an electric motor that accelerates the flywheel to store energy or as a generator that produces electricity from the energy stored in the flywheel. Flywheels can discharge their power either slowly or quickly, allowing them to serve as backup power systems for low-power applications or as short-term power quality support for high-power applications. They are relatively temperature insensitive, take up little space, have lower maintenance requirements than batteries, and are durable. You are also able to predict with accuracy the amount of energy remaining in the system.
High-speed flywheels (25,000 rpm to 80,000 rpm), use composite rotors made with carbon-fiber materials. The rotors have a very high strength-to-density ratio and rotate in a vacuum chamber to minimize aerodynamic losses. The use of superconducting electromagnetic bearings can virtually eliminate energy losses through friction. Once charged, the discs will spin, without losing any energy to friction, almost indefinitely. A power electronic rectifier/inverter, necessary to correct the large volume swing at the generator terminals over the speed range, is connected at the output of the generator. At this point, the energy is injected into the DC link circuit of the system.
Low-speed flywheels (2,000 rpm to 10,000 rpm) are made of steel and use conventional bearings with magnetic support, which is used to support the rotor weight. Only a partial vacuum is necessary to maintain low frictional losses. These losses, combined with the excitation losses, result in a float charge that's roughly double that of typical high-speed storage systems. Low-speed flywheels use either a single-stage insulated gate bipolar transistor or a rectifier/inverter for power transfer, which can be used to correct for changing rotational frequency. Some flywheels operate with an induction or mechanical coupling instead of this DC interface, which must compensate for this through power electronics that act as a 3-phase exciter winding in the generator.
Sidebar: Building a Better Compressed-Air Energy Storage System
Compressed-air energy storage (CAES) is a hybrid storage/power-production system. Off-peak electricity is used to power a motor/generator that drives compressors to force air into an underground storage reservoir, such as a rock cavern or abandoned mine. When the demand for electric power peaks, the process is reversed. The compressed air is returned to the surface, heated by natural gas in combustors, and run through high-pressure and low-pressure expanders to power the motor/generator/turbine to produce electricity.
As of mid-2004, only two CAES plants have been completed and placed in operation. One limitation is that the system requires an unused empty salt dome, aquifer, or abandoned mine. In addition, the system is not self-contained but depends on a pipeline to supply natural gas for the combustion chamber. Because of these disadvantages, installation has only been attempted on utility-scale generating plants, typically over 100MW.
The thermal and compressed-air storage (TACAS) system overrides the limitations of the CAES system by combining thermal and flywheel technology. Compressed air and thermal energy drive an expansion turbine for long-duration outages while a small flywheel system gives instantaneous response to step loads and short outages. Compressed air is stored in conventional gas cylinders or pressure vessels at 4500 psi or more. The air is then routed through a thermal storage unit, a stainless steel core maintained at 1300°F, to transfer heat to the to compressed air. The heated air spins a turbine-alternator to produce electric power. Air that exits this turbine is below room temperature and can be used to cool the protected load. Tanks that store the compressed air become cold during discharge, absorbing heat from the ambient environment and ultimately converting the heat into additional backup power. The system is self-contained — no connection to a combustible gas source — and has a 20-year life expectancy.
Sidebar: Building a Better Supercapacitor
Supercapacitors, also known as ultracapacitors, are electrochemical storage devices that work like large versions of common electrical capacitors. Unlike batteries, they store energy in an electrostatic field rather than in chemical form. The energy is stored as a charge or concentration of electrons on the surface of a material. They are capable of fast charges and discharges and can typically be recharged hundreds of thousands of times. Their power is only available for a short duration, and their self-discharge rate is much higher than with batteries. In power systems, they are most likely used as bridging power sources for UPS systems.
To charge the ultracapacitor, you add electrons to the carbon layer. One of the electrons becomes positively charged, and one becomes negatively charged. The ions migrate through the separator to their respective spots. When you discharge the capacitor, they all move back the other way.
The devices consist of two electrodes made from a conductive alloy containing metal oxides, carbons, and conducting polymers; a separator; and an electrolyte, which can be solid state, organic, or aqueous, depending on the type of application. The separator is a membrane that allows mobility to charged ions but restricts electronic conductance. This is rolled into a double layer called a “jellyroll,” and then placed into an enclosure. The size of the jellyroll depends on how much energy is going to be stored in the device. If the ultracapacitor cell measures 3,000 farads, the jellyroll will be approximately 4 inches long by 3 inches in diameter.
Like a battery, the operating voltage of a cell is about 1.3V to 2.5V. By combining parallel strings of supercapacitors, hundreds of volts can be achieved. Because of their chemical inertness, expected performance vs. reasonable cost, and technical maturity, the device shows promise for industrial applications.