A new power-transistor technology, advances in RF MEMS, and oscillator innovations are among the technologies that could disrupt current design approaches.
Despite the constant innovation in the microwave industry, many products are rooted in the same technologies. Gallium arsenide (GaAs), for example, is at the heart of many of today's advanced amplifiers and other active components. Amidst this seeming status quo, however, lurk new innovations that could change the way that many products are made. These "disruptive technologies" can take the form of brand-new, completely novel approaches that have never been seen or reported. Or they may comprise an older technology that fell out of favor, but now has been reconfigured or re-invented to meet the needs of current and nextgeneration technologies.
The RF power-transistor architecture that was just introduced by HVVi Semiconductors (www.hvvi.com) falls under this "oldie but goodie" category, as it leveragesbut at the same time revolutionizesthe discrete silicon power transistor. No major advances have been made in silicon RF-transistor design in 15+ years. This architecture, which targets radar/avionics designs, is based on a High Frequency, High Voltage Vertical Field Effect Transistor (HVVFET). That architecture promises to deliver superior frequency bandwidth, voltage, and power levels when compared to current bipolar and LDMOS technologies.
The company already debuted its first three products based on this new HVVFET architecture. They target high-power, pulsed-RF applications in the L-band like IFF, TCAS, TACAN, and Mode-S. All three transistors operate over supply voltages from 24 to 48 V. For pulsed applications in the L-band from 1030 to 1090 MHz, the HVV1011-300 operates at 48 V (see figure). It delivers more than 300 W of pulsed output power while providing 15 dB of gain and 48 percent efficiency for typical performance. It features a pulse width of 50 s and a pulse period of 1 ms. Thanks to the vertical device structure used in the HVVFET architecture, it also boasts high reliability and ruggedness. The device is specified to withstand load mismatches equal to a 20:1 voltage standing wave ratio (VSWR) at all phase angles under full rated output power.
Its siblings, the HVV1214-025 and HVV1214-100, are enhancement-mode RF transistors for L-band pulsed-radar applications in the 1.2-to-1.4-GHz frequency range. Both devices operate off a 48-V supply and produce 25 and 100 W, respectively. Under test signal conditions characterized by a pulse width of 200 s and a pulse duty cycle of 10 percent, the HVV1214-025 typically offers 17.5 dB gain while the HVV1214-100 provides 19.5 dB gain. Both transistors are capable of withstanding an output load mismatch corresponding to a 20:1 VSWR at rated output power and nominal operating voltage across the entire frequency band of operation.
From a system's perspective, the HVVFET's gain, efficiency, and power density could allow designers to eliminate amplification stages in power amplifiers (PAs), reduce parts count, and shrink printed-circuit-board (PCB) space requirements. At the same time, the technology's highly rated ruggedness allows radar and avionics designers to eliminate bulky and costly isolators. The architecture also flaunts a scalable wafer process. It can support higher power levels with the same layout and design just by increasing the size of the die.
Another new technology is promising to rid mainstream electronics of their last non-CMOS componentthe quartzbased resonator. Like every part of the electronics industry, the timing segment has been striving for more integrated and smaller package solutions. The CMOS Harmonic Oscillator (CHO) technology from Mobius MicroSystems (www.mobiusmicro.com) promises to satisfy this need while achieving accuracy levels that are beyond the reach of CMOSonly oscillators. The first product to use CHO technology is the MM8511. This spread-spectrum clock-generator integrated circuit (IC) operates from a 3.3-V supply. It can be used as a fully integrated clock generator with output frequencies that range from 100s of kHz to 100s of MHz. The initial products will be factory programmed at common interface frequencies in the 10-to-100-MHz range. The device offers a wide selection of spread-spectrum-modulation percentages from 0 to 6 percent.
To lower electromagnetic interference (EMI), the MM8511 builds upon the standard spread-spectrum clock-generation (SSCG) technique. By utilizing SSCG, EMI can be reduced by up to 15 dB in the critical frequency bands deterministically. It can therefore reduce engineering iterations to ensure EMI compliancy. Unlike conventional SSCG implementations, however, the MM8511 CHO replaces both the quartz-crystal and phase-lockedloop (PLL) ICs with a monolithic CMOS die. This die generates spread-spectrum clocks without the need for an external resonator. The elimination of the quartz crystal improves reliability, removes bulky packages from the system, and lowers the bill-of-materials (BOM) count. By doing away with the PLL, the IC improves performance by reducing clock jitter. It also lowers phase noise and power consumption.
CMOS circuit technology also is at the heart of the oscillator technology from Discera (www.discera.com). Yet this company melds CMOS with microelectromechanical- systems (MEMS) resonator technology. Thanks to its MEMS roots, Discera's MOS1 oscillator promises to provide superior temperature stability and standby current consumption. The MOS1 family offers a frequency range of 1 to 125 MHz in four product ranges: 1 to 10, 10 to 40, 40 to 80, and 80 to 125 MHz. At the highest frequency range of 125 MHz, the MOS1 current consumption is typically only about 11 mA. At a commonly used frequency of 27 MHz, the typical current consumption is about 4.5 mA. In addition, all family members feature a standby current consumption of less than 1 A. The MOS1 oscillator also delivers impressive stability over environmental conditions. The family is offered in three industry-standard temperature ranges: 0 to +70C, 20 to +70C, and 40 to +85C. The product is specified over the normal industrystandard shock (5000g, 0.3 ms, sine) and vibration (20g, 10 to 2000 Hz) levels. Yet it also has been successfully tested to 30,000g shock.
MEMS was heralded as the "next wave" in electronic design long before actual products were being created. Current RF MEMS efforts focus largely on passive components, such as capacitors, inductors, resonators, and switches. The MEMS approach provides these components with low loss, high Q factor, high linearity, and good power handling. In addition, they can be fabricated in fewer lithography steps than monolithic microwave integrated circuits (MMICs). Despite these benefits, only a handful of companies have come to market with actual RF MEMS products. Yet MEMS continues to be a hot area. EPCOS (www.epcos.com), for example, just acquired the MEMS business from NXP. In MEMS sensors, probably the longest-running player is Analog Devices (www.analog.com). Over the past year or so, TeraVicta (www.teravicta.com) has announced several microwave switches. In addition, MEMS innovations continue to hail from MEMSCAP (www.memscap.com) and STMicroelectronics (www.st.com).
MEMS is an example of a technology that garnered a lot of headlines early in its existence. Yet many of these disruptive technologies seem to come out of nowhere. One can often predict their onset, however, by keeping a close eye on industry journals. Often, work on such technologies seems to start almost simultaneously at research and development laboratories across the globe. A current example is frequencyselective materials. When used to make antennas, for example, such materials allow one component to be "tuned" to different frequencies. In doing so, they permit designers to implement fewer antennas in a design. Similarly, the same components may be used to meet the different frequency requirements of varying standards throughout the world.
At their core, frequency-selective materials must have a resonant bulk material that exhibits conductive properties at one or more frequencies within a single or multiple conductivity bandwidths. At every other frequency, it should exhibit non-conductive properties. This bulk material can be used in antennas as well as devices like frequency converters. In addition, frequency-selective approaches have been at the heart of many recent innovations in radar-absorbing materials.
These examples of disruptive technologies show how often designers are just one innovation away from designing through completely different approaches. Due to the vastly different needs of applications across the industry, such a myriad of techniques and products is very much in demand. As such, new techniques and approaches will continue to be spawned. Yet it is up to the fortunetellers to say which ones will truly disrupt current design techniques to dominate the industry.