This article reviews the arrival of commercial high-temperature superconducting (HTS) motors in the market. American Superconductor is concentrating its motor efforts on ship propulsion. The company has a contract with the US Navy’s Office of Naval Research to design and develop propulsion motors up to 33,500 hp. The big advantage of a superconducting motor aboard a ship is its small size, which frees up valuable square footage in the hull for the many other components needed in battle. Because superconducting motors will be about half the weight of their conventional counterparts, the efficiencies an assembly line brings to manufacturing suddenly open for many of them. Lighter, smaller designs also will translate to time saved in testing. Many of the technologies used in the 200-hp machine transferred to the 1000-hp unit, and many new techniques developed as well.
Reports from several motor makers this past summer are adding weight to the notion that commercial high-temperature superconducting motors could be with us shortly.
In July, American Superconductor Inc. of Westborough, Mass., ran a 5,000-horsepower synchronous machine. In Erlangen, Germany, Siemens started a 400-kilowatt machine, calling it Europe’s first superconducting motor. Meanwhile, Rockwell Automation of Greenville, S.C., demonstrated a 1,600-hp HTS machine for the U.S. Secretary of Transportation, Norman Mineta.
As outlined here in April 2000, American Superconductor is concentrating its motor efforts on ship propulsion. The company has a contract with the U.S. Navy’s Office of Naval Research to design and develop propulsion motors up to 33,500 hp. The big advantage of a superconducting motor aboard a ship is its small size, which frees up valuable square footage in the hull for the many other components needed in battle.
Indeed, comparing the latest 5,000-hp superconducting motor with a more common 5,000-hp machine, a first reaction is that the new motor excises a hefty lump of metal from the great bulk of the mature technology’s model. According to American Superconductor’s Max Mulholland, who manages motor and generator development, the new HTS motor uses about a quarter the copper of a conventional rotor and saves nearly the same 75 percent of copper in its stator. In rough numbers, a 5,000-hp HTS motor would use 350 versus 1,400 pounds of copper in the rotor and 900 versus 3,350 pounds of copper in the stator, he said.
For ships, then, the weight and size advantage of an HTS machine makes sense. In land-based industries, space is cheaper. Mulholland conceded that marketing HTS motors there on the basis of smaller size and weight for a given horsepower would be a tough sell.
He pointed out, however, that a few industries would find more power from a smaller footprint a decided advantage:
Mining is one, where big machines need to be lowered through narrow shafts; metal forming is another, where mounting a lithe motor overhead is preferred to sinking a 10,000-hp behemoth in a pit; and transportation is a third, where most vehicles must bring along their own motive power wherever they go. For these industries, superconducting machines make sense also.
For big machines—1,000 hp and up—a cost saving doesn’t necessarily come from the reduction in motor size and the resulting decrease in material going into a machine. Conventional materials are far less expensive than exotic superconducting wire. And though it takes fewer feet of superconductor than copper to build the rotor, the relative cost of the materials doesn’t yield any savings there either.
It’s the way in which superconducting motors will be fabricated that will potentially reduce costs, Mulholland said. Today, big motors must be “built in place,” he explained. A motor assembly remains resident on a particular patch of factory floor throughout construction. Workers go to it, rather than having it come to them.
Test and Retest
After it’s built, a big motor must be tested, too. Because of size and weight limitations on rail and truck transport, the heavy machines are usually taken apart and shipped as components. Once the pieces arrive on site, a large machine must be reassembled under field conditions. Then, it must be tested a second time in an environment that’s probably a lot less controlled than the factory.
Because superconducting motors will be about half the weight of their conventional counterparts, the efficiencies an assembly line brings to manufacturing suddenly open to many of them. Lighter, smaller designs also will translate to time saved in testing, Mulholland said. Such a motor, tested at the factory, could ship fully assembled. Further testing on site would lessen also, if it didn’t vanish completely.
Then there are the costs saved during operation. Because motors are already quite efficient, the expected 1 to 2 percent increase in efficiency might not sound like much, Mulholland agreed. But considered another way, a superconducting motor can cut the losses because of inefficiency by a quarter to a half.
Look at a typical 5,000-hp machine operating 24 hours a day, Mulholland suggested. Rounding up its energy use to 4,000 kW to simplify the arithmetic, that motor uses close to 100,000 kWh every day. Improving its efficiency by 1 percent saves practically 1,000 kWh a day. If electricity costs 10 cents per kWh, a superconducting machine running full time could save almost $36,500 annually—about a quarter of the price of a conventional synchronous machine, Mulholland said.
Yet, the wire fabrication process still has a way to go to even begin competing commercially with ordinary windings, according to Georg Nerowski, a project manager at Siemens who headed the company’s recent two-year superconductor motor effort. Its 400-kW (about 550-hp) synchronous HTS machine used superconducting tape from Copenhagen-based Nordic Superconductor Technologies A/S.
Nerowski estimated that the expense of superconducting material would have to be trimmed to 10 percent of current cost before some of the advantages of assembly-line manufacturing would begin shining through. He’s a bit cautious about foretelling the day when every single motor 5,000 hp and higher will be swapped for a superconducting machine. His company builds a lot of motors, too—about 5,000 a year from 200 kW to 5 MW in size.
Along with the other manufacturers, Siemens expects the principal applications of these fancy machines will be in ship propulsion, generators, and fast synchronous motors, the last because of smaller rotor diameters that could allow higher rpm. Conventional asynchronous machines are comparatively simple beasts whose grip on the world of rotation could be difficult to loosen, he said.
According to Department of Energy estimates, motors 1,000 hp or larger consume fully 25 percent or more of the electricity produced in the United States. Surprising, isn’t it, that the government targets efficiencies in consumer appliances under its Energy Star initiative, while ignoring incentives for high-horsepower motor users to move them into more efficient machines, Mulholland said.
It may be early for that, however. After all, the U.S. government has been promoting work on superconducting motors for a number of years, through both the Navy and the Superconductivity Partnership Initiative, or SPI, a program sponsored by the Department of Energy and various industry interests.
Formed a year after the discovery of high-temperature superconductivity in 1987, American Superconductor has only recently produced HTS wire in what Eric Snit-gen called a commercial product, “something that can be used and is robust enough to be sent into factories.”
Snitgen, who manages American Superconductor’s wire program, said that a tremendous level of engineering had to go into producing superconducting wire that could withstand the bumps of daily use as well as ordinary wire can. BSCCO, the brittle bismuth-based superconducting ceramic, doesn’t bend readily.
Although superconducting wire looks ordinary enough, it takes quite a few steps beyond mere wire drawing to ready superconducting materials for winding into coils. First, superconducting powder is packed into a silver matrix. Then strengtheners encase the matrix.
“All your superconductivity is in the powder,” Snitgen said. “Silver provides substance.” It allows the material to be formed. A thin lamination of stainless steel—thinner than cellophane tape—adds strength.
Into the Machine
Siemens’ Nerowski said his company built its machine to demonstrate some of the mechanical and electrical concepts behind superconducting motors. The company wanted to evaluate claims of reduced size, weight, and losses. Apart from differences in output, which dictated different voltage levels for their respective armature windings, the Siemens machine and the American Superconductor model are quite alike, in principle at least, Nerowski said.
The Siemens motor fits in a standard frame, whose shaft height is 350 mm (measured from the ground to the shaft centerline). Similarly, the American Superconductor 5,000-hp motor fits in a NEMA frame.
In both machines, superconducting wire forms the rotor coils only. Each rotor assembly resides in a thermoslike, vacuum-sealed cryostat. The field winding temperatures are maintained between 25 and 35 K through a commercial refrigeration system that pumps a chilled gas through the rotor by way of a cold head.
According to Mulholland, the American Superconductor machine uses the same kind of cold head that magnetic resonance imagers use in their superconducting versions. The cold head permits the refrigerator and coolant circulators to remain outside the moving rotor.
An electromagnetic shield located beyond the cryostat protects the rotor windings from stray fields produced in the stator. It absorbs energy from any short-circuits that might occur in the stator winding. And, it dampens torsional vibration.
The stator itself uses no iron teeth as a conventional armature does, due to an increase in the strength of the magnetic field produced by the superconducting windings. This construction feature, called an air core, prevents the saturation of iron that would normally occur in a field of 1.7 tesla—about twice the strength of the magnetic field in a conventional motor. This feature further lightens the stator, but requires special considerations when it comes to supporting and cooling the armature.
After putting its motor through a battery of standard tests for synchronous machines (heat run, short-circuited armature winding), Siemens ran the motor through special tests, Nerowski said. One test evaluated the motor’s response to oscillations that occur as a rotor loses its mechanical or electrical load. Researchers confirmed that the machine remained quite stable, he said. American Superconductor reported similar stability in its larger machine.
American Superconductor supplied the HTS coils, their supports, and the refrigeration for the 1,000-hp superconducting motor built under the government’s SPI program. Managed by Rockwell Automation, the 1,000-hp motor project sought to maximize operational efficiency, Mulholland said. In contrast, American Superconductor’s intent with its 5,000-hp machine was to lower manufacturing cost and to build a machine as small as possible. As a result, stators, rotor coils, refrigerators, exciters, and torque transfers are different between the two models, Mulholland said.
According to David Driscoll, manager of research for the Rockwell superconducting motor group, efficiency and size are both important targets for a superconducting motor competing in industrial markets.
“Even though you take out some of the manufacturing costs, you have a rotor that’s more complicated,” Driscoll said. “You can’t build anything that’s cheaper than a conventional induction motor. Even if you take out a lot of the size, you are still going to have a hard time competing with that,” he said. The systems that must accompany a superconducting motor add complexity, too, he said.
In considering potential markets for its super-efficient machines, Rockwell expected them to command higher prices but felt that any cost difference would be made up quickly in energy savings. Initial applications would be those with stringent size requirements, such as propulsion for ships. With advances in technology, superconducting motors are expected to work their way into other industries.
MEs Fix Troubles
Most of the problems with superconducting motors are mechanical, Driscoll said. His staff is heavy with Ph.D.s in mechanical engineering and lighter on electrical engineering doctorates.
Since demonstrating the first superconducting motor in 1996 (a 200-hp unit) and rerating the 1,000-hp design to 1,600 hp, Driscoll and his staff have been busy applying test results to their next machine—a 5,000-hp model. Among the investigative areas, the engineers are examining critical components such as refrigeration systems, coil cooling, and brushless exciters, as well as investigating technology for rotor manufacturing.
Many of the technologies used in the 200-hp machine transferred to the 1,000-hp unit, and many new techniques developed as well. For instance, the 1,000-hp machine used a composite torque tube that acts as a thermal break between the room and the cryogenic environment. “That was the first time we know of where anyone used a composite shaft in an electric motor,” Driscoll said.
The 1,000-hp machine achieved a number of superconducting firsts, at least for Rockwell: The designers nested composite torque tubes just inside the windings to shorten up the rotor. They used a brushless exciter system. They also used a closed-cycle refrigeration system. (The economics of the 200-hp machine didn’t allow that—instead, a gaseous helium bath was vented to the atmosphere.)
The engineers installed ferrofluidic seals in the transfer coupling—another first, Driscoll said. Rockwell engineers developed their own optical telemetry system. By employing 10 cryogenic temperature sensors outside the rotor and a number of voltage sensors inside, the engineers can monitor the condition of the cold space as well as the rotor while the motor is running.
It’s not likely that a production machine will be instrumented in that manner. The optical telemetry system is strictly for research, Driscoll said.
Yet, as research into the phenomenon of superconducting continues, its practical application to rotating equipment draws nearer every day.
The time may not be far off when a ship or train will move under superconducting’s long-awaited power.