he Power Sources Manufacturers Association (PSMA) is a not-for-profit, industry organization whose membership includes representatives of power supply and component manufacturers as well as universities, consultants, and other types of companies with interests in the power electronics field. The PSMA's stated purpose is "to enhance the stature and reputation of its members and their products, improve their knowledge of technological and other developments related to power sources, and educate the entire electronics industry, plus academia, as well as government and industry agencies as to the importance of, and relevant applications for, all types of power sources and conversion devices."

In pursuit of these goals, the association has organized a number of working committees that address technical issues of concern to PSMA members in areas such as lead-free regulatory compliance, energy efficiency, semiconductor technology, packaging and other topics. This article will focus on the work being done by the PSMA's Magnetics Committee, including its work in reprinting classic texts, organization of special sessions at industry conferences, sponsoring of power magnetics research, and collaboration with other groups.

After reading about the Magnetics Committee's efforts, some engineers may wish to get involved with its activities. While PSMA membership is encouraged, it is not required to participate in the technical committee work or to attend the monthly call-in meetings. So if this overview of the Magnetics Committee's projects sparks your interest, don't hesitate to contact the PSMA to obtain further information and to learn how you can play a part.

History of the Committee

The Magnetics Committee has been active for several years. Initially, it focused on providing a check list for designers of power supplies who must specify magnetic components when dealing with their magnetics suppliers.
Many of the classic magnetics textbooks that were used in the U.S. went out of print. The PSMA Magnetics Committee identified some of the leading texts and was able to acquire the rights to reprint selected ones. As a result the PSMA has been able to make these classic texts available to designers and students around the world. PSMA prints these and offers them at a minimal price as a service to the industry.

Among the classis books being reprinted by the PSMA Magnetics Committee are these titles:

• Transformers for Electronic Circuits; Grossner, Nathan R.
  
• Applications of Magnetism; Watson, J.Kenneth
  
• Modern Dc-to-Dc Switchmode Power Converter Circuits; Severns, Rudolf P. and Bloom, Gordon
  
• Soft Ferrites-Properties and Applications, 2nd edition; Snelling, Eric C.
  
• Electronic Transformers and Circuits; Lee, Reuben; Wilson, Leo; Carter, Charles E.

These books are available through the PSMA or How2Power's bookstore.
Meanwhile, the committee continues its effort to identify other classic books on magnetics design and applications that could be reprinted as a benefit to the power sources industry. If anyone has a favorite classic book that is out of print that he or she thinks should be made available, please let us know. The subject does not have to be magnetics.

Recently, the committee has identified some applied research projects to sponsor. Depending on the results, this research could provide useful tools for designers concerned with incorporating magnetic components in the design and development of new power products. More details on this sponsored research is given below.

Industry Sessions

The PSMA Magnetics Committee sponsors an Industry Session at APEC every other year, alternating with the PSMA Capacitor Committee. At APEC 2012, Steve Carlsen and Matt Wilkowski chaired an industry session titled, "Trends for AC Power Loss of High Frequency Power Magnetics." Within that session, the following presentations were given:

• "Ferrite Watt Loss Testing: Common Manufacturing Practices," by Zack Cataldi, Magnetics, Pittsburgh, PA.
  
• "Measurement and Modeling of Core Loss in Powder Core Materials," by Chris Oliver, Micrometals, Anaheim, CA.
  
• "A Series of New High Frequency Core Loss Measurement Methods," by Mingkai Mu, Virginia Polytechnic Institute and State University--Center for Power Electronics Systems, Blacksburg, VA.
  
• "Overview of Core Loss Prediction and Measurement Techniques for Non-Sinusoidal Waveforms," by Charlie Sullivan, Dartmouth University, Hanover, NH.
  
• "Core Loss Modeling of Inductive Components Employed in Power Electronic Systems," by Jonas Muehlethaler, ETH Zurich, Zurich, Switzerland.
  
• "Coupled Electromagnetic and Thermal Analysis of Ferrite Core Electronic Planar Transformer," by Mark Christini, Ansys, Cannonsburg, PA.
  
• "Development of Low AC Power Loss Multi Layer Inductor (Today and Future)" by Kiyohisa Yamauchi, FDK, Shizuoka, Japan.
  
• "Expert Design and Empirical Test Strategies for Practical Transformer Development," Victor Quinn, RAF Tabtronics, Geneseo, NY.
  

The PSMA Magnetics Committee is now planning the next magnetics-focused industry session for the upcoming APEC 2014, which is to be held March 16 to 20, 2014 in Fort Worth, TX. So far, the following topics have been proposed.

• "Transformers, chip scale to utility scale." In this presentation, the unique challenges of transformer design for different size and purpose could be compared and contrasted.
  
• "Are Magnetics Catching Up with SiC & GaN?" with topics like:
  
  • Higher temperature
    
  • Higher frequency switching, up to 20 MHz
    
  • Higher voltages

Planning is in the early stages, so suggestions for topics from readers are welcome. We want very much to identify topics that are important to the power electronics community.

Special Projects

The PSMA encourages its technical committees to sponsor special projects that will benefit its members. Although some of these projects fall outside the scope of the Magnetics Committee, you may find them interesting nonetheless. Noteworthy is a special project sponsored by the PSMA Packaging Committee on PSoC, Power Supply on Chip. Another major project is the Energy Efficiency Database, sponsored by the PSMA Energy Efficiency Committee. Yet another major project is the Power Technology Roadmap, sponsored by the PSMA Roadmap Committee. They published an update this year.

Of course, there are special projects specific to the PSMA's Magnetics Committee such as the reprinting of classic books mentioned above. Another one began in 2010, when the PSMA Magnetics Committee sponsored the first of several projects on magnetic core loss at Dartmouth under the direction of Dr. Charles Sullivan.

Core-Loss Projects at Dartmouth

The PSMA core-loss studies have their roots in my frustration with the traditional core-loss estimations. Magnetic parameters for core-loss calculations are not familiar to many electrical designers. There is more standardization now on SI units, but many data sheets and application notes use other units, with strange conversion factors. Is it MKS, cgs, rationalized MKS, Gaussian or SI? The conversion factor for oersteds to A/M is 103/4π, once you have figured out what is an oersted (named for Hans Christian Ørsted) and what's an ampere-turn per meter.

There is no need for all this confusion, as everything can be expressed in volts, seconds and amperes. The likelihood of making errors when converting back and forth using unfamiliar units is significant. It can be daunting even to try.
Manufacturer's core loss data is notoriously inaccurate. It is not unusual to see ruler straight lines for functions that have significant curvature, with very large errors at the ends. This is sometimes justified because the lotto-lot variation is significant. The geometry of a core is a significant factor, so data taken on a "standard core" may not relate well to a large EE core.

When specifying a specific magnetic component, there is no need to use dimensional factors, any more than one would buy a resistor using the resistivity and the dimensions of its core. If core loss data is taken for a specific component, all of the dimensional parameters and their effects can be factored in. Given volts and seconds for the excitation, the loss can be read directly in watts. The data should be taken close to the intended operating point of the component.

Most core-loss data is taken with sine-wave excitation. This was fine when most transformers were for line frequency or audio, but now power converters use high-frequency, low-duty-ratio rectangular-wave excitation. Conversion formulas for low-duty-ratio excitation are notoriously complex, hard to use, and inaccurate. We can do better, and taking data with rectangular-wave excitation is a good start.

There are no good SPICE models for simulating core loss. While this topic is not part of any of the present coreloss studies, it is hoped that the knowledge from these studies will lead to an improved SPICE model. A modest start has begun.

Pilot Project

The Pilot Project was approved by the PSMA and a purchase order was issued to Dartmouth in the spring of 2009. Data was taken on one ferrite core and one powdered-metal core to explore a new proposed method of quantifying core loss using square-wave excitation. The composite-waveform hypothesis was partly validated, and shown to be an improvement over other approximations, both for accuracy and for ease of use.

An unexpected discovery of the Pilot Project was that the core loss per cycle is increased with increased off-time between the excitation pulses.

Phase II Project

The Phase II Project was approved by the PSMA and a purchase order was issued to Dartmouth in the spring of 2010. The Phase II project had two principle objectives:

• To test the composite-waveform hypothesis on a variety of cores of different materials, with emphasis on ensuring that the off-time loss phenomenon was not just a test-rig or test-procedure artifact.
• To test a core that had been drilled through with sense windings installed, to see if flux migration may contribute to the off-time loss phenomenon.
  

A byproduct of the Phase II project is a new SPICE model for core loss.

Phase III Project

The Phase III project has not yet been completed. Its purpose is to test core losses at low duty ratio and with a new winding configuration.

However, analysis of preliminary data suggests an improved Stienmetz-like equation. Another test showed that if a single core is replaced with a string of smaller cores with the same total volume, the core losses are reduced at higher frequency. This was one test on one specimen, so, while tantalizing, the conclusions are decidedly preliminary.

The core-loss studies performed at Dartmouth will be the subject of a future article in this column.

Phase IV Core-Loss Project

The core-loss studies to date at Dartmouth were greatly facilitated because the university had test equipment left over from previous work that could be adapted easily for our projects. However, continued work requires test waveforms and analysis at much higher frequencies and higher power.

It was originally thought that continuing the work would require an investment in advanced test equipment, an endeavor that likely would be beyond the means of the PSMA.

More recently, we realized that testing at higher power with fast, low-duty-ratio pulses was routine in the semiconductor industry. Therefore, we have begun talks with several manufacturers of GaN and SiC semiconductors, as advancing the science of high-frequency magnetics is of mutual interest.

Planning for the Phase IV research is just beginning, so anyone who has suggestions or would like to participate is welcome to join. In addition, any company who would like to show their support for this project is encouraged to become a member of the PSMA.

Coordination with Other Organizations

The PSMA cooperates with a number of other organizations. For example, the APEC conferences are sponsored by PSMA, PELS, IEEE and IAS. Another case of collaboration, PSMA recently held a joint workshop with EPRI, "Are You Smart Enough for the Smart Grid?"

There is also some coordination of efforts with respect to the development of core-loss standards. The PSMA Magnetics Committee is working with PELS and the TTA to keep them abreast of developments in our core-loss research. We attended the PELS Electronic Transformers Technical Thrust Committee meetings in New Orleans in November 2012 and at APEC 2013 in Long Beach. While the PSMA does not participate in standards writing directly, we do provide technical input.

About The Author

Edward Herbert is a member of the PSMA's Board of Directors, co-chairman of the PSMA's Magnetics Committee and co-chairman of the PSMA's Energy Efficiency Committee. Over the years, Ed has held a variety of positions in industry, working as a design engineer, a project engineer, an engineering supervisor, and as an engineering manager. Since 1985, he has been independent, promoting patented technology for license.

Ed has been issued 54 patents with several more pending. Of interest to this forum, Ed is the inventor of the matrix transformer, a novel arrangement of the windings and cores that is particularly useful for high-frequency transformers, especially for high-current, low-voltage applications. The matrix transformer patents have all expired. However, the latest generation of this technology concentrates on minimizing parasitic impedance and its effects, as those can lead to increased losses in other components, resulting in a poor overall design.

Ed holds a Bachelor of Engineering degree in electrical engineering from Yale University.

Editor's Note: This article was first published in the April 2013 issue of How2Power Today (www.how2power.com/newsletters).

If you or anyone in your company is interested in getting on the distribution list for future issues of PSMA UPDATE, please send e-mail to: power@psma.com. Be sure to include your name and the name of your company.