Spinning into control: high-tech reincarnations of an ancient way of storing energy.

Departing trains at a rail station could someday get their initial oomph for free, each time saving the equivalent of several days' worth of electricity usage by an average U.S. household. The trains would rely on a concept already used in today's hybrid gas-electric ears: reuse of energy stored while braking to a stop. But while hybrid ears stockpile energy in massive arrays of batteries, the heart of the hybrid train might be a deceptively low-tech device--a flywheel.

A flywheel stores the energy that was used to make it spin, and it retains that energy as long as the wheel is free to turn. Slow down the flywheel, and you can draw some of that energy back out.

Potters have taken advantage of flywheels for thousands of years. Since ancient times, they've shaped their bowls and cups on a wheel that they've turned by pushing on a pedal. The wheel's stored energy, or rotational inertia, keeps it turning at a roughly constant speed, despite the unevenness of the pedaling.

Engineers at the University of Texas in Austin are now experimenting with an improved flywheel that can store enough energy to get a standing passenger train up to cruise speed. Although such a wheel in a locomotive may weigh 5,000 pounds and its rim may move at the speed of sound, it would work on the same principle as the pedal-driven pottery wheel.

To store energy, either a mechanical system or alternating magnetic fields make a flywheel turn faster. To release energy, the systems slow it down.

Batteries are the most widely used way of storing energy and are the staple of hybrid vehicles of all kinds. But batteries are slow to recharge and typically can't give back their energy in quick bursts. After a few years, they lose their capacity to hold a charge and, because they contain toxic metals, can end up as hazardous waste.

Flywheels, their advocates say, can pack more energy than batteries of comparable weight. Flywheels can also last decades with little or no maintenance and are environmentally sound.

For decades, some engineers have fought a lonely battle as they've tried to promote flywheels for energy storage. Now, flywheels may be slowly starting to find wider acceptance--and practical applications. Several companies in the United States and Europe are developing flywheel-based hybrid buses and trams. Flywheels are also being east in an important role in the post-fossil fuel economy. They might be used to stabilize the output of solar, wind, and other energy sources, thereby reducing the risk of blackouts.

FREEWHEELING TRAVEL Interest in flywheels has come and gone in waves over the past several decades, as have investor interest and research funding. In the 1950s, cities were interested in clean mass transit that could replace trams and electric buses without needing tracks or suspended wires. A small Swiss company tried to commercialize what it called the Gyrobus. This vehicle's only source of power was a steel flywheel that was spun up while the bus docked briefly at specially equipped stops. But the Gyrobus was heavy, and its range was limited to a few kilometers, so it never reached mass production.

For decades, steel was the material of choice in flywheel engineering because of its high density. Other things being equal, denser materials store more mechanical energy. But density can be a mixed blessing. The outer rims of high-density wheels must withstand intense centrifugal forces. Engineers often saw their steel flywheels self-destruct. So, they had to restrict the top speed of the wheel rim to about 50 meters per second.

With the advent in the 1970s of light but strong carbon-based composite materials such as Kevlar, engineers realized that lighter could be better. The polymer fibers of composite materials make them several times as sturdy as steel. Their rims can move at more than 1,000 m/s. The high speed more than offsets the reduced density of composites. Doubling the spinning speed of a flywheel quadruples its stored energy, while doubling its density only doubles that energy.

To reduce energy losses from friction, engineers began enclosing flywheels in vacuum containers and suspending them on electromagnetic, rather than mechanical, bearings. And by embedding magnets in such flywheels, designers could arrange for their speeds to be controlled by magnetic systems, so that the flywheels would have no physical contact with the rest of the world.

In the mid-1990s, some start-up companies proposed flywheel designs to replace batteries or other power sources in electric cars. The flywheels came up to speed in a matter of minutes, rather than the hours needed to recharge a battery. But car manufacturers, with little interest in electric cars, didn't invest in the unproved technology.

Also-in the 1990s, NASA became interested in using flywheels to power the International Space Station. Once every hour and a half, the station's orbit brings it into Earth's shade, where its solar panels are useless. For about 30 minutes, the station relies on batteries. But frequent charge-and-discharge cycles wear out the batteries within a few years.

At NASA's request, engineers at the University of Texas in Austin started developing a flywheel pack that could store twice as much energy as batteries of the same weight and last the entire life of the station without maintenance. The project, however, fell victim to budget cuts.

Nevertheless, the Austin team achieved remarkable flywheel spinning speeds. Its carbon-composite prototype reached more than 50,000 rotations per minute and a rim speed of l,400 m/s.

More recently, the team has had locomotives in mind. In late April, inside a concrete bunker that provides safety during high-energy experiments, the researchers began testing their largest flywheel yet It's a cylinder 1.5 m in diameter and 1.2 m tall, spinning on a vertical axis. It is designed to store 133 kilowatt-hours (kWh) of energy, which the team claims as a record for carbon-composite flywheels. That energy would take a train from a standing start up to cruising speed.

The flywheel is made up of concentric shells, each one prestressed in a different way to withstand centrifugal forces that vary greatly from the axle to the rim. The new machine rests inside a 2.1-m-tall steel containment vessel that houses the electromagnetic suspension and a magnetic system that can spin the flywheel at up to 15,000 rotations per minute.

The team estimates that a flywheel-based hybrid locomotive could attain a 15 percent increase in efficiency on a route such as New York to Boston. "But the best payoff would be for commuter rail," says University of Texas physicist Robert Hebner. Although his team hasn't calculated the probable savings, stop-and-go travel in conventional vehicles wastes a lot of energy in braking that could be retained by a flywheel, he says.

Flywheels might also make a comeback in smaller-scale transit. Several teams in Europe and the United States--including the one at Austin--have recently proposed modern analogs of the Gyrobus. In Rotterdam, the Netherlands, a company called Centre for Concepts in Mechatronics developed a flywheel-powered hybrid bus and tested it in passenger service last year, says project manager Rien Beije.

The prototype bus incorporated a small car engine whose only task was to keep the flywheel spinning. The engine ran at the constant speed at which its fuel efficiency was optimal. The flywheel stored up to 3 kWh, ran on conventional ball bearings, but was kept within a vacuum. When needed, the flywheel could supply bursts of 300 kilowatts, the equivalent of about 400 horsepower.

The bus "ran like a Porsche, Beije says, and had 35 percent better mileage than a comparable-size conventional bus.

The company now has a contract with French engineering giant Alstom to develop a wireless tram that could recharge at stops.

BLACKOUT PROTECTION Meanwhile, some companies are now producing larger, stationary flywheels intended to make electric power more reliable. The electric power grid is "a subject that involves little passion until the system fails; says Ruth Howes, a physicist at Marquette University in Milwaukee.

The Electric Power Research Institute in Palo Alto, Calif., estimates that hundreds of brief power outages cause the U.S. economy to lose at least $120 billion a year, for example, by causing computer users to lose data. A single major disruption, such as the one that left the northeastern United States and Ontario in the dark on Aug. 14, 2003, can cost extra billions of dollars.

The electric-power grid operates on a careful balance of supply S and demand. Sudden imbalances in one place create disruptions that propagate over the grid by affecting the current's frequency or voltage. Increased reliance on inherently erratic power, such as solar or wind, will only make matters worse, according to a committee summoned by the American Physical Society and chaired by Howes. The committee is about to release a report highlighting the problem and listing flywheels as one of the possible solutions.

Flywheels are one of several types of devices, known as power-quality units, that can dampen changes in current's frequency or voltage by injecting extra juice into the grid. Such units would respond to fluctuations within fractions of a second and could prevent major emergencies such as the 2003 blackout.

"That entire power outage could have been suspended if the grid had had power-quality units, says Jim Fiske, a senior engineer at LaunchPoint Technologies, a company based in Goleta, Calif.

Some U.S. companies are already producing commercial flywheels for power -quality applications. Beacon Power, based in Wilmington, Mass., has designed and installed a 9-kWh model for large telecom clients that need stable power in remote locations. More recently, the company gained approval from grid operators in California and New York after testing its flywheel on both states' grids.

Beacon is now developing larger, 25-kWh flywheels meant to be installed in arrays of as many as 200. Such arrays would provide 20-megawatt bursts of power when needed to stabilize a grid's frequency.

Amory Lovins, chief scientist at the Rocky Mountain Institute, an energy- conservation think tank in Snowmass, Colo., agrees that a quick injection of power into the grid could have prevented the 2003 blackout. "It would have taken a few hundred extra megawatts, at most," he says. However, Lovins says that the best way to avoid a major power outage is to fine-tune the supply side by enlisting the help of large, industrial customers. "They will take demand off the grid [on short notice] if you pay them to do soS he says.

But for preventing smaller, more frequent disruptions, flywheels offer a better solution, maintains Beacon spokesperson Gene Hunt.

Meanwhile, LaunchPoint is looking into an alternative design for flywheels. Even when made of composite materials, flywheels such as those produced by Beacon or the University of Texas team are still limited in size by centrifugal forces. Those forces increase from axle to rim, and so tend to pull apart concentric layers.

Countering those forces adds complexity and cost, says LaunchPoint engineer Fiske. "Each separate rim has to have the right structural characteristics," he says. The LaunchPoint team proposes to remove everything except the outer rim, leaving just a hollow cylinder. The stress in the radial direction is minimized, and the wheel can be much larger and weigh up to tens of tons, Fiske says.

LaunchPoint's hollow design is further simplified by putting the entire electromagnetic bearing system on the inner side, instead of the outer side or the bottom of the rim. Fiske's team is building a small prototype with a capacity of 3 kWh. He says that the design will be easy to scale up, reaching a storage capability of at least 1 megawatt-hour--six times the Austin team's record.

Recent infusions of research money from the U.S. Department of Energy, the National Science Foundation, and NASA are helping small companies and university research labs bring flywheel technology to maturity.

Consumers are just beginning to get used to hybrid ears. A new generation of flywheels might bring the hybrid concept into systems ranging from from trains to the nation's entire electric grid.