The transmission grid that delivers electricity from power plants is a vital piece of America’s infrastructure. It is also good at hiding its flaws. People may notice the towers and wires marching across the landscape or the local substations that step down the voltage so electricity can be distributed to homes and businesses, but the transmission grid does not show congestion like highways do or flooding like burst water mains do. Nevertheless, the grid needs a major upgrade. If the U.S. is going to switch from dirty fossil fuels to cleaner, more renewable wind and solar power—or even nuclear—the transmission system must be vastly expanded to reach the remote deserts and high plains where the sun shines most and the wind blows hardest. Furthermore, if the country wants to protect itself against increasingly large blackouts, which cost tens of billions of dollars or more a year, it needs to modernize the grid as well.

So how do we build this supergrid? After years of debate, most engineers agree that a modern overlay should be added on top of the old, piecemeal, overtaxed system, creating a backbone that has greater capacity by using higher voltages and reaching more remote locations. The Obama administration’s 2009 stimulus package allocated $6.5 billion in credit for federal agencies to build power lines and $2 billion in loan guarantees for private companies, so money is available to get started. Constructing the supergrid will require several big technical steps, and one political.

Build, Baby, Build
The first step is simply to erect more transmission lines, especially extending from potential hotbeds of renewable energy to growing cities where demand is met now with coal-fired plants. New lines would also help regional utilities sell surplus power, when they have it, to utilities that are far away.

Large-scale construction is long overdue. As the country’s Investor in Chief Barack Obama said about the current system in October 2009: “Just imagine what transportation was like in this country back in the 1920s and 1930s before the Interstate Highway System was built. It was a tangled maze of poorly maintained back roads that were rarely the fastest or the most efficient way to get from point A to point B.”

During the past decade, only a little more than 1,000 miles of high-voltage transmission lines were built each year. But a National Renewable Energy Laboratory study released in January concluded that supplying 20 percent of the country’s power with wind would require 22,700 miles of new interstate electrical highways, on top of the more than 160,000 miles of existing high-voltage lines. The sight of workers erecting power lines would be more common than the sight of crews building roads.

In addition to connecting renewables, more lines would solve a vexing surplus problem. In a growing number of markets today, even when demand for electricity is low, certain power plants must run to keep voltage stable across the system, yet there is no demand for the actual power they are producing. At night, when winds are often high, there may be no place to send the electricity they create. In these situations, some transmission managers, such as the California Independent System Operator, are forced to pay power generators such as wind farms to cut their output. And if that still outpaces demand, “you pay people to take [the power],” laments Yakout Mansour, CEO of the California operator.

This imbalance can make clean, renewable energy awfully expensive. Forcing wind turbines to stop producing when the wind is blowing can quickly make them uneconomical. Electric highways can dilute the surplus, sending it to customers who do need power but are far away. More lines can also help spread out voltage surges and dips across a larger area of suppliers and consumers, so the fluctuations can be absorbed without creating dangerous voltage spikes or meddlesome blackouts or brownouts.

More lines would also make storage facilities for mass energy more feasible. Surplus wind energy at night could be stored by any number of technologies that can generate power during the next day when needed: big batteries, flywheels, compressed air chambers, water pumped uphill so it can later fall through turbines, molten salt tanks heated to later drive steam turbines, and so on. But the power needs to reach those facilities in the first place.

Pump It Up
More transmission lines will better connect generators and ­users. But transmitting power at higher voltages—the second step to a supergrid—will reduce losses along the wires, saving money and lessening the footprint of power lines running over hill and dale.

Power is lost along a transmission line primarily as heat, yet losses drop significantly as voltage rises. James A. Muntz, manager of transmission at Northeast Utilities, which wants to import more electricity from Canada, has calculated the large loss reductions for a 100-mile line loaded with 800 megawatts (MW), roughly the output of a large coal plant. If the line is operating at 345 kilovolts (kV)—the level used along many backbone wires today—19.8 MW of power is lost. At 765 kV, the highest voltage operated in the U.S. (though not widely used), only 3.45 MW is lost—about one sixth of the losses at 345 kV. For a 1,100-kV line, the loss is a mere 1.91 MW. The Soviet Union once operated a 1,150-kV line, Japan has a similar line and China is building several.

In addition to saving utilities money daily, higher voltages could help regional planners reduce the land and construction costs required for power lines. One 765-kV line can carry as much energy as six 345-kV lines, according to Michael Heyeck, a transmission executive at American Electric Power, which is the largest U.S. operator of such lines. A 765-kV line requires a footprint along the ground that is 200 feet wide, but the six 345-kV lines require 900 feet. The standard 765-kV power line tower is 135 to 150 feet high, however, significantly taller than 345-kV towers, which typically reach 110 to 125. As voltage rises, the lines need more clearance above the ground, yet the public generally looks on higher towers as more visually intrusive.

In January 2008 American Electric Power and the U.S. Department of Energy unveiled one possible plan to lay a nationwide 765-kV backbone over the existing transmission system—the way interstate highways overlay local roads—to greatly expand the grid’s capacity and lower its losses. The backbone would require 22,000 miles of 765-kV lines, of which 3,000 miles already exist. The network would cost about $60 billion. But billions of dollars could be saved every year because of significantly lower losses and because expensive local power could more readily be displaced with inexpensive power from places the grid does not now reach.

Go Direct
To further reduce losses, engineers recommend that lines along the most heavily traveled corridors use direct current, instead of the standard alternating current supplied to virtually all homes and businesses. Muntz calculates that the same 100-mile line loaded with the same 800 MW, but operating at 500 kV of direct current, would lose 3.82 MW, less than half the alternating-current losses at the same voltage. Push the line up to 800 kV, and losses drop to 1.5 MW, also less than half for a 765-kV alternating-current line.

Direct current is desirable for point-to-point tran­smission, with no stops in between. It is already used between hydroelectric dams in northern Quebec and New England and between dams on Oregon’s Columbia River and southern California. In these cases, direct current was selected because it is efficient and is controllable. Alternating current follows the path of least resistance, buzzing along random wires like water on a mountaintop trickling down various streams to a pool at the base. A direct-current line is like a pipe from top to bottom, with a pump that can be adjusted in real time.

Control is important because of the balkanized nature of the power grid, made of hundreds of local networks with separate owners. When a seller sends electricity to a buyer far away, the power will flow along whichever wires it chooses, potentially causing overloads in various places between the end points that have no commercial relation to the deal.

Direct-current cables have been opened in the past few years between the New Jersey coast and the southern side of New York’s Long Island and across Long Island Sound to Connecticut. In both cases, finding an acceptable underwater route was much easier than it would have been across the heavily congested metropolitan areas, and the lines provided express power flows. The 53-mile Trans Bay Cable was recently opened along the bottom of San Francisco Bay, carrying 400 MW. Control is so smooth that utility officials closed old, relatively dirty generating stations in San Francisco that had been operated to improve the voltage and frequency stability of the area’s alternating-current networks.

Despite the advantages, direct-current lines are worthwhile only if installed over long distances. That is because special converter stations are needed at each end to change alternating current to direct current and back again. The conversion requires massive electronics that eat up around 1 percent of the electricity at each end. According to Andrew Phillips, director of transmission at the Electric Power Research Institute, with declining costs, the break-even point has moved down to about 300 to 350 miles, compared with 500 miles 15 years ago.

If configured smartly, long direct-current lines could form a backbone across the continent that would differ from American Electric Power’s 765-kV plan, which employs direct current sparingly. The National Renewable Energy Laboratory study calls for 10 massive, 800-kV east-west direct-current connections from the Plains states to the Atlantic coast, although it did not specify routes. At a conference on building Midwest wind farms to supply the East Coast, Dale Osborn, transmission technical director of the Midwest Independent System Operator, said that direct current is the only technology that would guarantee that the power would get to exactly where it was needed. The complication is that tapping into a line at a midpoint, the equivalent of a highway interchange, is extremely expensive.

Knit the Nation Together
More long-distance transmission lines, at higher voltages, whether alternating or direct current, could form a supergrid that strengthens and extends the existing transmission system. Such lines would have to be built at a scale greater than planning has traditionally been practiced, however. The challenge is particularly acute with direct-current wires: an alternating-current line can be lengthened in increments, like extending a road a few miles at a time, but a direct-current line is like a bridge, with a fixed beginning and end.

Transmission has almost always been built piecemeal, within the territory of one utility or two neighboring utilities. The siting of new lines and obtaining rights-of-way always face bureaucratic hurdles and usually public opposition. A third challenge arises nationally, because the Lower 48 states are divided into three giant power grids: the Eastern Interconnection from the Rockies eastward, the Western Interconnection from the Rockies westward, and Texas. The three grids have largely functioned as independent islands for decades, and the eastern network is also subdivided like a jigsaw puzzle into regional pieces.

In an effort to modernize, regional operators in the Eastern Interconnection unveiled a plan in 2009 for a systemwide upgrade that would enable wind power to meet 20 percent of the grid’s energy needs by 2024. The plan called for 15,000 miles of overlaid transmission lines, half of which would be direct current. The map did not specify routes, but the lines could run along existing utility rights-of-way or along railroad lines or even highways.

The plan got stymied in part over how cost would be apportioned. One option would be “merchant” lines, the equivalent of toll roads built by private companies, a few of which exist today. But this arrangement works only when the buyer and seller can be precisely linked, meaning, only for direct current. An option for alternating-current lines is to split the cost among the generators and consumers who will be served; however, some regional transmission organizations that have tried this scheme have had long arguments about the formula. The Southwest Power Pool proposed a cost-allocation system, approved by the Federal Energy Regulatory Commission in June, under which higher-voltage lines are treated as highways and costs are divided among all utilities in the area. Lower-voltage lines are treated as byways; like local streets, their costs are borne locally. Costs for lines with intermediate voltage would be shared.

But that arrangement will not work for direct current, because the lines provide no benefit for anybody who is not at either end. A line that began in the Dakotas and ended in Chicago would be useless to Minnesota, Wisconsin and Iowa. For such a line, the cost of transmission might be built into the cost of electricity, a technique that Hydro Quebec is trying in a new line that will serve New England.

The troubled plan for the Eastern Interconnection finally fell apart when New York and New England states claimed that the plan had a bias toward tapping wind energy on the Great Plains and sending it to the eastern seaboard. The eastern states said the plan would preclude developing wind resources off the Atlantic coast, and they walked out.

A different attempt to integrate the three giant power grids is under way in New Mexico, in a spot close to where the three large interconnections touch. The region also happens to have plentiful wind and solar resources.

The three systems are not connected now, because their alternating currents are not synchronized. In each grid, the electrons alternate direction 60 times a second, at precisely the same moment, like Rockettes dancing in a chorus line. But the three sets of currents do not dance together; they are timed to different drummers.

Swapping energy among them requires switching the alternating current from one region to direct current, tying it to an adjacent region, then switching it to alternating current at the correct synchrony. Eight direct-current ties exist among the three interconnects, but they can transfer only a modest 1,500 MW of power, the equivalent of two large coal plants.

A private venture called Tres Amigas proposes a single transfer station in Clovis, N.M., that would move power among all three interconnects on a larger scale. Silicon-based power electronics—not the fingernail-size chips in a computer but hunks of semiconductor the size of a stack of dinner plates—would chop the flow of alternating current into tiny pieces and reassemble it as direct current. The current would then move through superconducting cables, with extremely low losses, to another terminal where more semiconductors would reassemble the power into alternating current. The transfer station, estimated to cost $1 billion, could handle 5,000 MW of power flows and could be expanded to 30,000 MW if more robust power electronics were devised. The huge tie would act as an anchor for the three systems, transferring power and counteracting voltage instabilities.

Tres Amigas would make money by charging for power transfers and possibly by facilitating a market for selling and buying power, as the New York Stock Exchange does for stocks. It could also sell its stabilization of voltage and frequency as a service.

Hurdles loom, though. For example, Texas is not now regulated by the Federal Energy Regulatory Commission, and companies there do not want to be. Yet the state could benefit. “Texas has built all this wind and is up against the wall” with not enough local customers to buy it, says Phillip G. Harris, chief executive of Tres Amigas and a former president of the nation’s largest independent system operator, PJM. Other locations with extensive renewable energy would have a way to ship it, too.

Action or Obstruction
One of the greatest hurdles facing a truly national supergrid is the geographical and financial scale. For the federal government to organize and fund it, as it did for interstate highways, a strong national mandate for renewable energy would probably be needed. Another pathway would be setting a dependable price on carbon-based fuels or carbon dioxide emissions that either creates a pot of money or gives renewable energy an edge, which could spur private-sector funding of a supergrid to deliver it.

At the moment, though, the prospects are uncertain. Transmission planning remains a state-level exercise, because states generally control land-use decisions. Without a strong push from a renewable energy mandate or a carbon charge, “there does not appear to be a great deal of stomach for a national plan for transmission,” concludes Jay Apt, executive director of Carnegie Mellon University’s Electricity Industry Center.

Indeed, in March a newly formed Coalition for Fair Transmission Policy, made up of giant investor-owned utilities, public power cooperatives, congressional Democrats and Republicans, and state energy officials, voiced opposition to a strong national electric grid, centrally planned and broadly financed, that would promote renewable energy. The group is trying to block the Federal Energy Regulatory Commission from approving a series of major transmission pathways from wind-rich areas in the middle of the continent to load centers across the nation. Other critics question whether the commission even has the authority to approve such lines. Senator Ron Wyden of Oregon, a coalition member, compares the proposed power lines to gas pipelines that would carry fuel between New York and northern California but might pass through Oregon, “with no direct benefit to the people in my state.” A modern network could benefit people in all states, however, by bringing more efficient, less expensive power to interconnected grids everywhere and reducing the likelihood of blackouts.