For four days in February 2012, three Chattanooga, Tenn., Local 175 members attached 99 rectangular-metal boxes to some transmission lines east of Knoxville that just might solve one of the thorniest problems of the American electrical grid.

Foreman Cody Young and his co-workers Ryan Swafford and Jeb Tenille worked about 20 miles of 161-kilovolt transmission lines owned by the Tennessee Valley Authority. The boxes, called "discrete series reactors," promised to do something very hard in a simple, durable way: steer electrical power away from overloaded transmission lines to underutilized parts of the grid. If there is a traffic jam in a city, drivers will take other routes. Electrons, though, always take the path of least resistance, which often means some wires in transmission grids are maxed out long before the grid as a whole is.

The result is the average grid operates at about 60 percent of capacity. Previous attempts to get electrons into less busy wires had been large, expensive and used lots of energy. The DSR gets all the energy it needs from the line it is clamped to and, autonomously or by remote control, can increase the impedance on a line, like squeezing a garden hose with a clamp.

To make installation easier, Smart Wire Grid, the Oakland-based startup that makes the DSRs, created a special attachment for the bucket trucks.

"We raised it up to the line, hinged the top over, and then closed it down with 20 bolts to secure it, and we went on to the next one," Young said.

Young and company got all 99 installed in less than four days and ahead of schedule.

If the DSRs work as expected, and that is still a question, carrying capacity on electrical grids could jump to 90 percent. At that level, a U.S. Department of Energy study found that up to 10 times more renewable energy production could be added to our current electrical grid without harming reliability.

"This is the kind of advance that can put 'smart' in the electrical grid and gives us a level of control we haven't had before," said Rob Manning, chief energy delivery officer and executive vice president for Tennessee Valley Authority.

Behind that moment in the hills, behind Smart Wire Grid, was a relatively small, virtually unknown government agency that invests in what President Obama called "high-risk, high-reward technologies" that tackle the nastiest problems in the American energy system: the Advanced Research Projects Administration.

Funding High Risk/High Reward Technologies

Originated under the Bush administration, President Obama launched ARPA-E in 2009.

"Energy is our generation's great project," Obama said, announcing $400 million in initial funding. "Today we face more complex challenges than we have ever faced before. In no area will innovation be more important than in the development of new technologies to produce, use and save energy."

It is often said that while electricity pioneer Thomas Edison, who died in 1931, wouldn't recognize most of the technology we use every day, he would have no trouble understanding our electrical grid. That may finally be changing. The energy department predicts hundreds of coal-fueled power plants will close down in the next few decades. Nearly a third of the transmission and distribution infrastructure is nearing or past retirement age.

At the same time, new technologies are changing the nature of the electrical system itself. More vehicles will get their power from the grid instead of gas. Hundreds of thousands of residential and commercial buildings will soon be sending power back out into the grid from on-site solar and wind power. A new generation of sensors and controllers will give utilities and regulators greater understanding and control over the country's more than 185,000 miles of transmission and distribution wires.

"The nation that leads the world in 21st century clean energy will be the nation that leads in the 21st century global economy. I believe America can and must be that nation," Obama said. "We will put in place the resources so that scientists can focus on this critical area. And I am confident that we will find a wellspring of creativity just waiting to be tapped by researchers … and entrepreneurs across our country."

ARPA-E is modeled after the Defense Advanced Research Projects Agency, a small arm of the Pentagon created after the Soviet launch of the first satellite, Sputnik. DARPA's job was to take back the lead in the space race and then win the Cold War, and in its history it was crucial in the creation of GPS, the stealth fighter and the Internet.

ARPA-E's job is to do the same for energy, by finding viable technologies with the potential to transform the American energy landscape but that are too early, too risky or too academic to attract regular investors. ARPA-E steps in where the market won't by getting inventors the resources — money, business expertise, lab space — to see if the idea can be translated into reality. Many of them will fail. The hope is that a few successes will be sufficient to create a radically altered world.

When he became ARPA-E's first director, Arun Majumdar said the agency's job was to go where other government programs, industry and even investors wouldn't if a technology had the potential to disrupt and transform fundamental problems.

"We think of ourselves as pre-venture funders, for [ideas] that are too risky for venture capitalists," Majumdar told Time magazine.

From the Lab to the Field

In 2005, professor Deepak Divan used new alloys of steel to create an ungainly prototype DSR in a lab at Georgia Tech. After six years of testing and refinement, Smart Wire Grid officers couldn't find the resources they needed to go into production from utilities, private investors or government agencies.

Smart Wire Grid applied for, and won, a $4.4 million grant from ARPA-E, one of the nearly 285 grants totaling $770 million it has given out in the last five years.

ARPA-E program managers worked with executives to set technical and business milestones. Grantees get reimbursed for expenses when they meet their goals. If they don't, they can be dropped from the program.

Mark Johnson has worked with a dozen companies as the director of the ARPA-E program focused on finding cost-efficient ways to store massive amounts of instantaneously available energy within electrical grids.

"Storage is the holy grail of building a 21st century electric energy system. Right now, generating and consuming, it's all simultaneous," Johnson said. "Storing energy at scale cheaply will be completely game changing."

Without storage, Johnson said, it won't be possible to get more than a quarter of energy generation from renewables. The most efficient way to store energy now is pumping water into reservoirs and then capturing it later using hydroelectric turbines. His goal was to find ideas that were as efficient and cheap, but that could be put anywhere in the country.

"If you do a good job rethinking the problem, it makes the previously unrealistic ideas more interesting," Johnson said.

Johnson was a professor at North Carolina State University and ran his own company before joining ARPA-E. His father, Rudy Johnson, was a member of Minneapolis Local 160 and worked at the Black Dog Power Station when Johnson was a boy. He says his father told him many times that without energy, you don't have an economy.

"I remember years when Dad worked Christmas Eve. He said, 'Someone has to keep the lights on,'" Johnson said. "I feel like I'm doing that too now."

Johnson started in the semiconductor industry in the '80s and said he sees parallels between those early days of microchips and today's energy technology.

"Those first chips were crude by our standards. That wasn't so long ago, and now they are incredibly sophisticated and powerful," Johnson said.

The most familiar way to store energy is a chemical battery, and half of the grantees are pushing the boundaries of electrochemistry. Beacon Power, though, created six-foot-tall fly-wheels that spin at 16,000 rpm, levitated by magnets inside a vacuum chamber. General Compression developed a system that uses underground caverns to store compressed air.

Johnson said nearly half his time is spent in the labs, helping the companies by keeping their focus on practical results.

"Very early we get to the applied part of applied research," Johnson said. "We ask them, 'Where will you take it if it works? Who are the customers? Who can manufacture it?'"

Johnson said ARPA-E connects companies with investors and utilities early in the process to make sure that the technology can be successfully integrated no matter how radical it might be.

"What you don't want is a widget that you finally get to work as a prototype but needs a whole other system to do any good," Johnson said. "It is not up to us how to commercialize the technology. We have to ask the questions though, and then help the company find their own answers."