Power electronics can slash electricity use 33%
Right, Fred Lee, director of the Center for Power Electronics Systems (CPES) and co-director Dushan Boroyevich, left.
A decade ago, Virginia Tech power electronics researchers adopted three verbs standardize, modularize, and integrate and developed a concept that not only triggered a multi-university, multi-million-dollar center, but also impacted billions of dollars of commercial technology and has the potential to slash U.S. electrical use by up to 33 percent.
Power electronics is the use of electronic circuits to convert and control electricity. The technology is used to process and convert raw, generated electricity into the voltage, wave form, and type (ac or dc) needed by today’s machines, motors, and electronic equipment.
In 1998, power electronics was a $60-billion industry worldwide that impacted another $570 billion in hardware electronic sales. Power electronics systems were typically custom-designed, non-standard units containing 300-400 electronic components. They were expensive and required long development times, plus there was little integration and fairly low reliability, according to Fred Lee, director of the Center for Power Electronics Systems (CPES).
The integrated power module
In the late ’90s, Virginia Tech’s power electronics researchers were tackling a problem for Intel, trying to boost the speed of the power processing so the Pentium II chip could achieve its design speed. The solution came from “taking the complicated solution and breaking it into small, modular pieces, then making it scalable,” Lee says. “We gave the idea the multi-phase voltage regulator module (VRM) to industry and today every microprocessor in the world is powered by the technology.”
The VRM, he says, helped to revitalize the U.S. power electronics industry. During the ’90s, power supply manufacturing was being moved offshore with the rest of electronics equipment. “The industry was bottom-line oriented, and there was little focus on research. Today, just the power modules for microprocessors alone comprise a couple billion-dollar industry.”
The VRM success triggered the Virginia Tech team to propose that power electronics technology for all-sized electrical applications, from aircraft to portable phones, could be made more efficient by integrating intelligent power electronics modules (IPEM). The concept met with approval among researchers and industry.
Long-term NSF support
In 1998, the Virginia Tech researchers teamed up with power electronics researchers at the University of Wisconsin-Madison and advanced power semiconductor researchers at Rensselaer Polytechnic Institute to form CPES, one of 20 Engineering Research Centers (ERC) sponsored by the National Science Foundation (NSF). They were joined by researchers at North Carolina A&T State University and the University of Puerto Rico-Mayagez, institutions with solid reputations in undergraduate engineering and power electronics-related research.
“This is an enabling technology that could have such high impact for both energy savings and U.S. industry, that it was a perfect candidate for long-term funding,” Lee explains. NSF ERC funding was a 10-year commitment, contributing $30 million to the effort. Another $47 million was contributed by industry partners and other funding agencies
The paradigm shift
Integrated Power Electronics Modules (IPEM) developed at CPES that could improve efficiency and reliability of future electric energy systems while decreasing costs. The top IPEM uses embedded power technology and the bottom uses flex power.
The paradigm shift from custom, complex systems to modular, integrated building blocks required technical advances in multiple areas simultaneously, from semiconductor adhesives, to system integration models. In pursuing these advances, CPES has generated more than 1700 technical papers, theses, and dissertations, 131 invention disclosures, 50 patents, and 331 licenses.
CPES demonstrated the first IPEM in 2000 and the concept moved quickly into the marketplace. Today, many power electronics firms manufacture some version of an IPEM, according to Lee. The concept has seen the most impact in three applications, he says: motor drives, electronic ballasts, and information technology. These are also three applications that can have the most impact on energy consumption.
Motor drives account for about 50 percent of the electricity use in the United States and have been a targeted application for IPEMs since the start, with a focus on variable-speed motors. Variable-speed drives that match the output to the load can often achieve an energy reduction of 30-35 percent. The larger motor drives that power factories and large equipment typically use variable speed technology. However, smaller applications, such as residential air conditioning and other home appliances, typically do not use variable-speed drives in this country, according to Lee.
“A constant-speed drive has a cheaper installation cost, even though it costs 30 percent more in energy use,” he explains. In Europe and Japan, where energy costs are higher, manufacturers offer variable-speed equipment, and Lee expects more offerings in the United States, now that IPEMs are reducing the initial purchase cost.
The development of motor drives with integrated power electronics is a major success for the CPES integration concept, Lee says. “The University of Wisconsin had the strongest reputation in U.S. motor drives research,” he explains. “Once they started working with the IPEM concept, the Wisconsin researchers developed a unique modular integrated approach in the motor. A motor was always a lump of iron,’ plus a separate electronics drive. Now, Wisconsin has an approach that integrates both parts together. They sliced the power electronics functions into pieces and integrated the drive module with the motor module. This is a huge advance.”
The country’s second biggest use of electricity 20 percent is in lighting. As the United States moves away from incandescent lighting, more lights will be using ballasts, which are used in fluorescent, LED, and other lighting. In 2002, CPES developed IPEM technology for electronic ballasts for high intensity discharge (HID) lamps and continues to develop technology to replace magnetic ballasts with power electronics.
“With our developments, the electronic ballast can now compete in cost with magnetic ballasts, and it is small enough to fit into the lamp fixture,” Lee says. “This will lead to even greater energy savings.” He cites Tech’s recent replacement of all the ballasts on campus with electronic devices and the resulting 30 percent savings in electricity usage.
Fred Wang, right, with student Luis Arnedo, is applying the modularized power electronics technology to high-power applications that range from tens of kW to megawatts of power. Applications include ships, factories, homes, and even microgrids.
Computing technology consumes between 10-15 percent of the country’s electricity. Starting with the success of the VRM, the CPES integrated module concept has taken off throughout the computing industry, Lee says. “We proved the concept with the VRM for microprocessors, then similar technology was developed for memory, servers, and other components. All the solutions are evolving around scalability, modularity, standardization, and integration. When you add in the movement to incorporating the technology in portable electronics, including cell phones, PDAs and GPS systems, that’s another multibillion dollar industry zeroing in on the same solution.”
The technology is currently being incorporated in the telecomm industry, he says. “It’s such an obvious solution, whether it is wireless or wired. When we came up with the multiphase idea, power supplies in switchboards were very customized with no standardization.” The old switchboards had converters that converted 48 volts to various other voltages, he explains. “Today, they are leveraging the VRM technologies and infrastructure developed for computers. They no longer need the highly customized and bulky 48 volts dc-dc converters often referred to as bricks’ in the switchboard, and the industry is saving 15 to 25 percent in the cost of telecomm power equipment.”
So many firms are using the concept that it is very hard to quantify the impact of the idea that triggered the technology, he says.
Power electronics education
The NSF funds an ERC for 10 years, averaging about 15 such centers at any one time. The goal of the ERC program is not just scientific and technological advancement, but also technology transfer and increasing the diversity of the scientific and engineering workforce. CPES has been cited as a model ERC for both education/outreach and for industrial collaboration and technology transfer.
The technology transfer is evident not just in the commercialization of so much technology based on the CPES concept, Lee says, but also through the 149 industrial internships and 250 technology transfer activities. “One of our own measures of success was to see the technology moved quickly to market,” he says.
When the ERC was first formed, few students across the country specialized in power electronics. In its 10 years of operation, CPES graduated 105 Ph.D. students and 181 master’s students. The center developed 15 new power electronics courses and 27 distance courses across the five universities, plus instituted an undergraduate minor to encourage undergraduates to enter the field.
Next up: sustainable buildings
The NSF ERC funding ended in 2008, but the five CPES universities plan to continue collaborating. “Almost unanimously, we all want to stick together and work on energy efficiency. We want to tackle alternative and energy efficiency solutions,” Lee says, “and we believe that future, sustainable buildings is an ideal application.”
Virginia Tech researchers are developing a dc-power distribution testbed, with a high-voltage (300 V) bus powering HVAC, simulated kitchen loads, and other major appliances; and a low-voltage (48 V) bus that coincides with telecommunications and computer loads. The system’s energy sources include a 3.5 kW wind-turbine generator, 5 kW PV solar panels, a lithium-ion battery bank for energy storage and a hybrid electric car with bidirectional energy flow.
University of Wisconsin researchers are developing a high-frequency (400 Hz or greater) ac power distribution testbed. With a high-frequency ac architecture, low-cost frequency converters can be used to efficiently power key motors in a home, such as the furnace fan, air conditioner compressor and fan, refrigerator, and heat pumps. The use of high-frequency systems may also reduce the cost of lighting converters while enabling the use of commercial-grade ac switchgear and protection devices.
Rensselaer researchers are developing an integrated energy system architecture that can provide and optimize the energy supply to portable entertainment and information appliances. The energy system module will be able to interact simultaneously with multiple energy sources, such as energy-harvesting MEMs, solar cells, and micro-fuel cells.
CPES researchers expect these efforts and others to have a major role in the country’s transition to sustainable development. “Our goal is to develop electronics that can reduce electricity use while enabling new technologies to evolve,” Lee concludes.
For more information, visit: www.cpes.vt.edu/.