Freescale RF Vice President: LDMOS will still be the technology of choice in many RF markets

JasonYoo

Freescale RF Vice President: LDMOS will still be the technology of choice in many RF markets [Copy link]

Since Freescale Semiconductor was spun off from Motorola in 2004, the company's RF division has undergone major changes in its markets and technologies. The RF division has made continuous progress in its core technology LDMOS (laterally diffused metal oxide semiconductor), while continuing to research GaAs and GaN technologies to meet emerging application needs. The division has entered the general amplifier market and is retaking this industrial market. They have also used their core technology in the WiMAX market and have gained significant market recognition. Microwaves & RF conducted an exclusive interview with Gavin Woods, vice president and general manager of the RF division, on the company's technology and market development plans.

MRF : From the perspective of technological achievements, such as power levels, efficiency and increased frequency, what is the mid-term and long-term roadmap for LDMOS compared to Freescale's compound semiconductor products?

Gavin Woods : Our LDMOS roadmap has always been to improve the key intrinsic performance, such as efficiency and gain. We have also been focusing on cost-effective device packaging, which is equally important as performance. As customers use new design methods to ensure savings in production costs and capital expenditures, Freescale will focus on device optimization to meet customer requirements. As for the long-term plan for LDMOS, we are sure that it will remain the technology of choice in many RF markets in the next few years. In the high-demand RF market, LDMOS has a clear leading position compared to compound semiconductor technologies such as GaAs and emerging technologies such as GaN.

Gavin Woods , Vice President

MRF : Is the maximum operating frequency the main limiting factor for LDMOS?

Gavin Woods : LDMOS is currently limited when you look at high frequency performance. By this I mean above 4 GHz. However, if you look back 10 years, the prevailing view was that LDMOS was good at 1 GHz and below, and that was true at the time. However, continued device engineering advances have pushed LDMOS operating frequencies to more than four times that frequency, and performance continues to improve. For example, in January we announced that LDMOS power devices can meet the requirements of WiMAX 3.8 GHz. WiMAX has very strict requirements for linearity and efficiency of power devices, and our new devices can meet these requirements like any other device on the market. For another example, they are significantly cheaper than GaAs devices. We have proven that we can continue to extend LDMOS to higher frequencies, and our device designers tell me, why don’t we push LDMOS to higher frequencies while maintaining good performance? There is no reason.

MRF : There is a lot of talk and research going on about the potential of gallium nitride (GaN) devices. Is Freescale developing this technology?

Gavin Woods : Yes. We are investing in GaN for potential future applications. We are positioned to be a leader in RF power, even though the required die technology is holding us back. The type of device that makes the most sense in a given application will give the customer the best cost/performance tradeoff, and that is what we want to offer the market. Right now, LDMOS is the primary focus because it offers the best cost/performance tradeoff. If GaN or other technologies can help us improve our leadership in the RF power market and help our customers design and develop more advanced products, then we will promote these technologies in our products.

MRF : In what applications do you think GaN will be competitive?

Gavin Woods : When it comes to generating RF power, GaN's main advantages over competing technologies come from its high saturation current density, high breakdown voltage and very high power density per cell gate width. These features enable GaN to efficiently transmit very high power levels over a wide bandwidth, and it has the potential to operate to 50 GHz, so communications, radar, satellite communications, electronic warfare (EW) and digital microwave broadcasting are all possible applications. The biggest challenge is growing high-quality GaN crystal epilayers on silicon or silicon carbide substrates. The lattice constant mismatch between the epilayer and the substrate can lead to defects, affecting quality and reliability, and this is obviously a topic that Freescale and other companies are passionately researching. Once the quality and reliability issues are resolved, the market demand for GaN will determine its cost. If there are large applications, its cost will naturally fall.

MRF Reporter : WiMAX is touted as a revolutionary development that complements or even replaces existing wireless communication technologies, replacing microwave links, T1 lines, free space optical propagation in cellular backhaul communications, and providing new options for home broadband Internet access such as DSL, cable TV and fiber. What is Freescale's positioning on WiMAX and how does the company seize this opportunity?

Gavin Woods : We are developing WiMAX devices at 2.3, 2.5 and 3.5 GHz and we assume that the market for these products will develop. If the market develops, we will extend our products to 5 GHz. We will treat it as the next big thing and will keep going until the customer tells us to stop. When we went to customers 6 to 9 months ago, there was very little interest in WiMAX. Now, we will not give up without a comprehensive review of what we have done, what we know about the market and what we are doing. So, the momentum is clearly gathering. We certainly believe that WiMAX can be a major player in fixed access applications and we have seen its vitality. Whether it will become a really big market depends almost entirely on whether it can provide the real flexibility defined in the IEEE 802.16e standard. If WiMAX can do this, the second hurdle is whether there will be a business case for the configuration of the WiMAX architecture and how it can cope with the next generation CDMA2000 EV-DO standard and HSDPA (High Speed Downlink Packet Access) and HSUPA (High Speed Uplink Packet Access) for WCDMA. Distributed deployments will emerge, which will make it a walk in the park to bring WiMAX-enabled devices. But the real potential problem is full mobility. If this passes the test, everyone in the wireless field will be interested in the wireless industry.

MRF : For device products, gain blocks appear to be a larger market for RF devices. Now that Freescale has entered this crowded arena, how will its products be accepted by the market? What challenges will it face?

Gavin Woods : We have responded correctly to the general purpose amplifier. It is obviously not easy to distinguish competitors who have been in this industry for many years and have a solid foundation. However, we entered the GPA industry because our customers asked us to do so, and they needed Freescale's reliability, quality and production scale. We made design improvements to the device, and we feel that this gives us a competitive advantage.

MRF reporter : Few media have paid attention to the industrial applications of RF power devices, such as magnetic resonance imaging (MRI), scientific systems, heating and other operating areas specified in the scope of industrial safety manuals. Undoubtedly, it is difficult for people to immediately think of their existence because the devices are deeply hidden in the products. These applications are scattered across a wide range of disciplines, and the industrial market is just a very real one among them. Can you talk about Freescale's motivation in developing the industrial market for RF power devices?

Gavin Woods : We have successfully re-entered the industrial market. We were very significant in the industrial RF power market six years ago, but we sold that business to focus on cellular infrastructure. Now we are re-entering the industrial market and bringing all the technologies we have developed for cellular architecture over the years - LDMOS and overmolded plastic packaging, which will allow us to deliver products with significantly better performance and lower cost than other providers.

MRF Reporter : What do you think of the current RF power device market?

Gavin Woods : There hasn’t been much innovation in the RF power market for 10 to 15 years, except for a few thermal improvements. There has only been a decline in year-over-year growth, but no significant cost reduction. We will take a step-function approach to performance and reduce product pricing. LDMOS technology and integral plastic packaging enable us to do this.

MRF reporter : Can you give some examples?

Gavin Woods : Okay. For example, our flagship industrial device, the MRF6V2150N, we are achieving significantly higher gain and efficiency at current levels. For example, the size of the driver you need will be much smaller because you get enough gain with the MRF6V2150N, which has a positive impact on cost, and the unit cost will be significantly lower than the current level.