Microscale Generator Yields Macroscale Power

May 2005
By Cheryl Lilie

Georgia Institute of Technology (Georgia Tech) researchers (l-r) Dr. Iulica Zana, David Arnold and Dr. Jin-Woo Park work with a dime-size generator they developed that can produce enough power to run a cellular telephone. The generator will be incorporated into a microengine that will power military electronics.
Mini device proves powerful enough to run military electronics.

Researchers are demonstrating that good things, in the form of useful amounts of power, can come in small packages. At the Georgia Institute of Technology in Atlanta, researchers have been able to produce power with a generator approximately the size of a dime. The device, called a microgenerator, is one aspect of a project to create a microengine that weighs less and lasts longer than batteries used by soldiers in the field today.

The microengine concept was developed approximately 10 years ago when the U.S. Army Research Laboratory, Adelphi, Maryland, began funding a research project at the Massachusetts Institute of Technology (MIT) in Cambridge. The Army was looking for a two-part system: a small gas turbine engine to convert chemical fuel—such as jet propulsion fuel-type 8, or JP8—into mechanical energy and a generator of the same size that would convert the mechanical energy into electric energy. The two components would need to be built simultaneously so that they could be integrated fully. Although MIT is still developing the engine component, the microgenerator portion of the project has created a device that can produce up to 1.1 watt of power—enough to run a cellular telephone—and researchers expect it to generate enough power to run a laptop computer soon.

The research team at the Georgia Institute of Technology (Georgia Tech) began working on the generator five years ago when asked by MIT to create a permanent magnetic generator prototype. MIT engineers originally planned to use an electrostatic generator because electrostatic devices are more common in microelectromechanical systems. However, they discovered that a permanent magnetic generator design could yield more power from the extremely small machine than could an electrostatic prototype of similar size.

The magnetic generator consists of two sections: the rotor and the stator. Because the engine is not being built at Georgia Tech, the microgenerator team needed to simulate the mechanical energy necessary to operate its part of the system. “For testing purposes, we are using compressed air to power a spindle similar to a dental drill,” explains Dr. David Arnold, postdoctoral fellow and a lead researcher on the microgenerator project at Georgia Tech. “We mount the rotor, which houses a magnet in a titanium casing, to the drill and then line it up and spin it over the stator.”

The stator has electroplated copper coils fabricated onto a chip in a process similar to microelectronics processing. Three pair of wires lead out of the coil to a power electronics circuit. “This three-phase configuration is the same as you would find in a large macroscale machine,” Arnold says.

The Georgia Tech group discovered that it could spin the rotors approximately 120,000 revolutions per minute (rpm). “To give you a reference,” Arnold offers, “a car engine at full throttle is only about 6,000 rpm. So we are spinning 20 times greater than that.”

Arnold notes that the reason the rotor can achieve this speed is that the machine is very small. As the generator is scaled down, the rotor can go faster and the generator can produce more power.

With this setup, the laboratory harnessed 1.1 watt of direct current power. “There have been other projects where microgenerators and micromotors have generated power, but they have always been in the milliwatt range,” Arnold reveals. “This amount of power is a breakthrough.”

The Georgia Tech team is looking at ways to increase the power output up to 10 or 20 watts. “The easiest way, or the most direct way, is to spin the magnet faster,” Arnold says. Speed and power have a quadratic relationship in this device: Doubling the speed increases the power output by a factor of four. The researchers are trying to extend the speed to 300,000 rpm, which would increase the power by a factor of 10.

But the high speed comes at a price. “The challenge is that the magnets we are using are very brittle, and at those high speeds, they tend to break apart and actually disintegrate under their own weight. Everything has to be very well balanced and precisely assembled,” Arnold says. The titanium housing helps solve that problem. Not only does it serve as a way to mount the magnet to the rotor shaft, but also it helps keep the magnet intact.

Another way to increase the power output is to change the type of magnet used. Arnold and his team selected a magnet made of samarium cobalt because it can withstand the high temperatures of a gas turbine engine without losing magnetization. “Our design goal for the cool section of the engine, where the generator would be, was 300 degrees Celsius,” Arnold says, but he admits that experimental elevated temperature tests have not yet been performed.

According to Arnold, the samarium cobalt magnet is the second best for the generator, and using the best magnet would increase power by approximately 20 percent. Enlarging the machine also would yield an increase in power. “We were working with these dimensions mainly to focus on the proposed size of MIT’s turbine,” he states. “But if we double the size of the machine, we would get four times the power.”

The 10-millimeter-size generator developed at Georgia Tech produces 1.1 watt of power. Researchers are looking for ways to increase the power output up to 20 watts in the near future.
For the Army Research Laboratory project, the size was limited to that of batteries soldiers use in the field today. According to Dr. Stuart Jacobson, a principal research engineer at MIT, the fuel tank size they are developing at MIT should fit the same form factor as batteries now available. “If you take the BA 5590 as an example, the main battery the Army uses for soldier portable applications,” Jacobson explains, “it is an LiSO2 [lithium-sulfur dioxide] primary battery that weighs 1 kilogram and is 127 millimeters x 112 millimeters x 62 millimeters. If we say we are drawing 16 watts continuously, a typical amount for future force use, the battery will last about 11 hours.” Jacobson reports that if a power source with a similar size and shape was filled with JP8 using 90 percent of the available volume and leaving room for the microengine and battery balance, the battery would work for 48 hours. In addition, because the fuel is less dense than the battery, the fueled system would weigh approximately 30 percent less than the present generation of Army batteries.

Until the gas turbine engine prototype is completed at MIT, the Georgia Tech team is looking into other sizes and applications for the microgenerator, including applications that do not require the turbine engine. Arnold says that the microgenerator could work anywhere that there is a source of fluidic flow. “You can imagine little propellers hanging off a small air vehicle. As this micro-airplane flies, it would turn a propeller and create power. Any situation where you have energy harvesting or energy gathering would work.”

Another application would be placing the generator on oil drilling machines to power sensors on the drill heads. The generator would use the compressed air from the spinning drill to power the sensors. However, one of Georgia Tech’s main goals now is to develop a stand-alone generator, eliminating the drill from the setup. In this configuration, the package would be the size of a matchbook, and compressed air would be blown across blades attached to the magnet. When the magnet spins, electric energy would result. Two terminals coming out of the generator would send the power to the electronic device.

But Arnold points out that originally the intent was, and still is, to integrate a microgenerator and a gas turbine engine. “There will be some hurdles to actually building them simultaneously all integrated, but that has always been our intent,” he says. “It will definitely be challenging, but it won’t be impossible.”

Although located more than 1,000 miles apart, the teams at Georgia Tech and MIT collaborate frequently. “At this point, our tests have not been with theirs [MIT’s], but we do work very closely with MIT,” Arnold says. He notes that MIT helped collaborate on the microgenerator design by sending students and professors to Georgia.


Web Resources
Georgia Tech Research Institute: www.gtri.gatech.edu
Massachusetts Institute of Technology Research: http://web.mit.edu/research
U.S. Army Research Laboratory: www.arl.army.mil


Enjoyed this article? SUBSCRIBE NOW to keep the content flowing.