Wide Band-Gap Semiconductors in the Big Easy

Jason Lomberg, North American Editor, PSD



It’s that time of year again – no, not the one where weird relatives stink up the joint and some fat guy zips around the world with magical caribou. I’m talking about APEC. The big one. The crown jewel. The World Series of power. And with it comes arguably the hottest topic in our fair industry – wide band-gap semiconductors.

First off, if you haven’t already, I highly recommend you check out our online article series, “The Power and Evolution of GaN” and “Are you SiC of Silicon?”, by EPC and UnitedSiC, respectively. Both were extremely popular, and we hope to do more projects like those in the near future.

And speaking of projects, GaN and SiC project to be huge at APEC (in the fair city of New Orleans), especially if the March issue is any indication. Our annual wide band-gap issue is like the hottest restaurant in town – everyone wants to get in.

First up, RS Components begins with the simple (but profound) observation that SiC power MOSFETs are nearly at the same price point as conventional silicon MOSFETs or IGBTs. Like most newer technologies, GaN and SiC require a higher upfront investment than legacy tech. But the gap, so to speak, is closing.

Efficient Power Conversion (EPC) discusses the process of testing GaN devices to failure. As Messrs. Alex Lidow, Robert Strittmatter, Shengke Zhang, and Alejandro Pozo state, the normal cycle of testing GaN designs at-or-near the limits specified in their data sheets is inadequate.

“By testing parts to the point of failure, an understanding of the amount of margin between the data sheet limits can be developed, but more importantly, an understanding of the intrinsic failure mechanisms can be found,” they state.

The inherent appeal of wide band-gap semiconductors – like many new technologies – is efficiency.  “GaN has a number of performance advantages including an electron mobility that is far greater than that of silicon (3.4eV vs. 1.1eV) which gives it the potential to conduct electrons more than 1000 times more efficiently than silicon,” notes Yong Ang with ON Semiconductor.

It’s also very possible that wide band-gap devices are more viable than we think. Transphorm's Philip Zuk believes that “the design techniques required to optimize GaN FETs are not as sensitive and complex as some would suggest,” and a lot of the agita surrounding GaN integration stems from the belief that all GaN is alike.

A functional smart grid is essential to our future (and the Internet of Things), and GaN has an important role to play. IGBT and silicon metal-oxide semiconductor field-effect transistor grid converters served the industry well in the past, but GaN versions promise new levels of performance and system cost.

And as Masoud Beheshti, with Texas Instruments, explains, “GaN-based multilevel converters can surpass not only traditional IGBTs but also SiC in efficiency, power density and solution cost.”

Finally, UnitedSiC, which previously authored our online SiC article series, ties silicon carbide in to the forthcoming electric vehicle revolution. UnitedSiC's Dr. Anup Bhalla mentions that “EVs following the ‘yellow brick road’ towards a green future in the emerald city are increasingly likely to have SiC in their power electronics, displacing older IGBT technology.”

Enjoy the March issue and the Applied Power Electronics Conference!

Best Regards,

Jason Lomberg

North American Editor, PSD