Technology Editor Chris De Martino spoke with John Palmour, CTO and Co-Founder of Wolfspeed about the company's intentions and his take on gallium-nitride-based technology.
|Download this article in .PDF format |
This file type includes high resolution graphics and schematics when applicable.
Earlier this year, Cree announced that its Power and RF division would become a standalone company. The new company, Wolfspeed, was introduced in September. John Palmour, who serves as Wolfspeed’s Chief Technology Officer, is one of Cree’s original founders. He spoke to Microwaves & RF about the Wolfspeed move, and provided his insight into RF technology—specifically gallium nitride.
CD: What prompted the decision to separate the Power and RF division of Cree into a standalone company?
JP: The issue was the fact that Cree had three separate components. The Power and RF division was the smallest, but it was also the most profitable. This division accounted for 8% of the company’s revenue. Because we believed it was undervalued, we did the “carve out” to unlock value we think is buried within Cree. With this move, we believe we can move faster, be more agile, and increase our stock value.
CD: What can customers expect to see from Wolfspeed as a standalone company?
JP: Our customers can expect to see more of the same, as we do not intend to dramatically change. The company’s focus will remain the same. We plan to continue to drive silicon-carbide (SiC) and gallium-nitride (GaN) technology forward. We believe we can now move even faster because of our dedicated resources.
CD: We’ve seen a huge explosion in GaN technology in recent years. How much of this technology’s potential is still yet to be realized?
JP: This can be answered from both a market standpoint and a technology standpoint. From a market standpoint, we believe that it’s still early. We are seeing much traction, as GaN is gaining a foothold in numerous applications. However, a significant amount of room still remains. It is our belief that the market will continue to go in the direction of GaN.
From a technology standpoint, we feel there are still things to be explored. We think the technology could enter higher frequencies. In addition, there are numerous ways it can be used at the same frequencies. This includes various amplifier topologies, as people determine the best way to work with this technology. We can discover the best-suited circuit topologies to take advantage of GaN for very high-efficiency power amplifiers.
CD: What applications do you see as the biggest beneficiaries of GaN technology?
JP: It comes down to bandwidth. GaN provides the advantage of wide bandwidth along with high power. This is why we’re seeing GaN utilized for 4G LTE applications. 5G will almost certainly require GaN technology. High-data-rate applications, such as satellite-communications (satcom), also benefit. Other beneficiaries include wideband military applications like jamming and radar. Ultimately, GaN can hit many applications by providing an unprecedented combination of power and bandwidth.
CD: What are some of the difficult high-power requirements you receive from customers?
JP: The difficult requirements are across the map. One challenge has been achieving cost requirements. Plastic packaging is one example, as customers require both high power and low cost. Additional challenges include the standard requirements, such as linearity and efficiency.
CD: Do you believe that, eventually, GaN will completely replace gallium arsenide (GaAs) and laterally diffused metal-oxide semiconductor (LDMOS)?
JP: I believe so for high-power amplifier applications. As for handsets, that is still a ways off. Although applications with wide bandwidth requirements will utilize GaN, LDMOS will probably not go away. GaAs is good for low-voltage requirements, but it, too, is becoming stressed because of bandwidth requirements.
CD: Ka-band has become a very popular frequency band for satcom. What can we expect to see from Wolfspeed to support Ka-band?
JP: Our focus has been directed toward Ku-band. Many of our released products target this satcom band. Our 40-V process, for example, allows us to achieve more power at this frequency band. We are currently working on a 0.15-μm process, as we have already conducted some early foundry work. I’d say that Ka-band is probably a year away. But much of our effort will continue to be focused on Ku-band.
CD: You mentioned a 0.15-μm process. What frequencies will be reached as a result?
JP: Our Ka-band (millimeter-wave) process is targeting applications through 44 GHz. We expect to begin offering foundry services followed by catalog packaged products using this process within the next year. This process will open up access to higher frequency markets for us, including higher-frequency satcom and backhaul, while offering a higher-performing transistor at lower frequencies.
CD: Have you seen instances of GaN-based power amplifiers replacing traveling-wave tubes (TWTs)?
JP: This is an ideal application for GaN. There is quite a bit of development happening in the industry across a number of common military and commercial bands, including the S-, C-, X-, and Ku-bands, as well as broader bandwidths, including 6 to 18 GHz and even 4 to 18 GHz.
CD: Do you expect to see GaN technology used increasingly more to design other components, such as low-noise amplifiers (LNAs)?
JP: Yes, there is a move across both military and commercial segments to improve dynamic range in receiver-based systems. With the increased power handling of GaN, LNAs can be manufactured that are able to accommodate much higher input powers than conventional LNAs made from traditional III-IV semiconductors. This, combined with the potential elimination of limiters, provides for higher dynamic range with similar system noise figures.