This Small Business Innovation Research (SBIR) Phase I project proposes to develop a new class of DC-DC converters by employing modern control methods to map nonlinear power converters such as the boost and buck-boost topologies to linear models resulting in a simple and reliable converter design. A novel and breakthrough control strategy is proposed that allows nonlinear DC-DC converters to behave linearly for the full output voltage range and respond quickly to load and source changes. The invention provides significantly better performance and supports a wider range of output voltage compared to existing commercial products and offers a cost competitive solution compared to existing design techniques by reducing the design effort. The methods can be employed for any nonlinear converter resulting in faster response, lower cost, and higher efficiency. The design is independent of desired output voltage and stabilizing gain, and is free of right-half-plane zero effects. The concepts will be demonstrated by simulating and designing boost and buck-boost converters using input-output linearization. Parameters in the control laws will be transformed to simple algorithms that can be implemented in a small foot-print application specific integrated circuit. Key applications such as solar systems and light emitting diode lighting will be targeted.

The broader impact/commercial potential of this project involves development of energy efficient power electronic systems that are becoming increasingly critical in a broad range of industrial, portable, consumer, military, and medical applications. Efficient power electronics that minimize idle power and improve power factor can reduce the loading on the grid. Existing applications that employ nonlinear converters such as boost and buck-boost topologies will benefit from elimination from the right-half-plane zero and stability problems, thus achieving higher bandwidth. This indirectly helps improve efficiency. Further, the proposed method allows these converters to be designed in a much simpler fashion thus reducing the design effort, time, and cost. Also, a single controller can be employed for a wide operating range, reducing inventory requirements as compared to present techniques that require different linear models and designs for different operating points. This will allow nonlinear converters to be employed in new portable electronics and other applications that use lower voltage batteries rather than employ buck converters with higher voltage batteries. The broader technical and commercialization vision involves efficient power management systems that are cost effective for manufacturability and deployment. These issues have been key limitations to date for large scale deployment of these converters.

Project Report

Purpose The purpose of this project was to demonstrate in practice a theory supporting a completely new way to design power converters for variable output environments that would enable them to be made more simply, with fewer parts, using existing off-the-shelf components, and yet possess greater capabilities. Background Power converters are ubiquitous devices present in almost all forms of electronics. Their function is to take the voltage from the electrical power source (either a battery or from an electrical outlet, for example) and then step up or step down that voltage to provide the right amount of power to each of the many parts in an electrical device. This is a $40 billion global industry and power converters are critical parts in everything from cell phones, computers, motors, to industrial equipment and electrical grids. Every year greater demands are being placed on power converters in terms of lowering their cost and increasing their capabilities. Additionally, the architecture and design behind power converters hasn’t changed significantly in the past 20 years. There have been improvements in the various subcomponents and parts used in power converters, but the overall method of designing them hasn’t varied greatly in decades. This has become a significant barrier to increasing the electrical efficiency of electronic devices. Technology The approach taken by Cirasys has been to design a next-generation "smart" converter that is able to rely on mathematical algorithms based in control systems theory and embedded in software in a processor chip to create a converter that is able to manage varying levels of output voltage using a greatly simplified design (compared to existing design means). The same single-controller design is used at each output voltage, something not possible using traditional design methods. Results Cirasys used the NSF SBIR Phase 1 grant to prove this unique and novel approach by creating two technology demonstration circuit board designs, a boost (which steps up voltage) power converter and a buck-boost (which steps up or down voltage) power converter. The boost converter has an input of 12 volts with an output range of 13-40 volts, and it is able to switch within this output range using only one controller. Voltages may be selected arbitrarily: any voltage point within the 13-40V range is possible. The buck-boost converter has an input of 12 volts and an output range of 4-26 volts, again arbitrarily selectable. These converters use unique and proprietary algorithms to conduct input-output linearization computations which allow them to quickly switch on the fly to the voltage level selected. The only other way to accomplish this through traditional design methods is by using a different model and control design for each different output voltage point, an expensive and complex approach. Experiments on these boards also indicate that they are able to provide a constant output voltage when provided a varying input voltage. Impact Cirasys was able demonstrate in actual hardware a new design approach that can use fewer subcomponents, thereby lowering the overall cost of a converter and increasing its electrical efficiency (compared to an existing "fixed voltage" design), and provide greater capabilities to electronic devices that operate in variable voltage environments. This has significant broader impacts for a number of devices in the marketplace. Solar panels and wind turbines have varying voltage inputs throughout the day – a power converter with this approach would be able to rapidly and economically accept all the many levels of incoming voltage and produce a constant output. Power amplifiers would be able to provide an almost infinite range of voltage outputs based on demand and need. There are numerous other applications for variable power converters, from power supplies, computer servers, hybrid and electric vehicles, motor control, industrial power factor correction, battery management, medical diagnostic imaging systems, high-efficiency "green" technologies, and a number of defense applications as well. Deliverables The products that have resulted from this NSF award include boost and buck-boost power converters that are technology demonstration units in a printed circuit board platform. The project was done with the collaboration of the University of Texas at Dallas and has given complete and thorough design and test experience for graduate students. A design for an integrated circuit version (ASIC chip version) of this new power converter architecture was also created. A large amount of data has been collected regarding electrical efficiency, design improvements, frequency switching rates, and many other elements and parameters of this new approach to power converters.

Agency
National Science Foundation (NSF)
Institute
Division of Industrial Innovation and Partnerships (IIP)
Type
Standard Grant (Standard)
Application #
1013815
Program Officer
Juan E. Figueroa
Project Start
Project End
Budget Start
2010-07-01
Budget End
2010-12-31
Support Year
Fiscal Year
2010
Total Cost
$149,934
Indirect Cost
Name
Cirasys, Inc.
Department
Type
DUNS #
City
Dallas
State
TX
Country
United States
Zip Code
75252