David B. Rutledge

Fully integrated CMOS power amplifier design using the distributed active-transformer architecture

A novel on-chip impedance matching and power-combining method, the distributed active transformer is presented. It combines several low-voltage push-pull amplifiers efficiently with their outputs in series to produce a larger output power while maintaining a 50-/spl Omega/ match. It also uses virtual ac grounds and magnetic couplings extensively to eliminate the need for any off-chip component, such as tuned bonding wires or external inductors. Furthermore, it desensitizes the operation of the amplifier to the inductance of bonding wires making the design more reproducible. To demonstrate the feasibility of this concept, a 2.4-GHz 2-W 2-V truly fully integrated power amplifier with 50-/spl Omega/ input and output matching has been fabricated using 0.35-/spl mu/m CMOS transistors. It achieves a power added efficiency (PAE) of 41 % at this power level. It can also produce 450 mW using a 1-V supply. Harmonic suppression is 64 dBc or better. This new topology makes possible a truly fully integrated watt-level gigahertz range low-voltage CMOS power amplifier for the first time.

Distributed active transformer-a new power-combining and impedance-transformation technique

In this paper, we compare the performance of the newly introduced distributed active transformer (DAT) structure to that of conventional on-chip impedance-transformations methods. Their fundamental power-efficiency limitations in the design of high-power fully integrated amplifiers in standard silicon process technologies are analyzed. The DAT is demonstrated to be an efficient impedance-transformation and power-combining method, which combines several low-voltage push-pull amplifiers in series by magnetic coupling. To demonstrate the validity of the new concept, a 2.4-GHz 1.9-W 2-V fully integrated power-amplifier achieving a power-added efficiency of 41% with 50-/spl Omega/ input and output matching has been fabricated using 0.35-/spl mu/m CMOS transistors.

The class-E/F family of ZVS switching amplifiers

A new family of switching amplifiers, each member having some of the features of both class E and inverse F, is introduced. These class-E/F amplifiers have class-E features such as incorporation of the transistor parasitic capacitance into the circuit, exact truly switching time-domain solutions, and allowance for zero-voltage-switching operation. Additionally, some number of harmonics may be tuned in the fashion of inverse class F in order to achieve more desirable voltage and current waveforms for improved performance. Operational waveforms for several implementations are presented, and efficiency estimates are compared to class-E.

A 100-MESFET planar grid oscillator

A 100-MESFET oscillator which gives 21 W of CW effective radiated power (ERP) with a 16-dB directivity and a 20% DC-to-RF conversion efficiency at 5 GHz is presented. The oscillator is a planar grid structure periodically loaded with transistors. The grid radiates and the devices combine quasi-optically and lock to each other. The oscillator can also be quasi-optically injection-locked to an external signal. The planar grid structure is very simple. All of the devices share the same bias, and they can be power and frequency tuned with a mirror behind the grid or dielectric slabs in front of it. An equivalent circuit for an infinite grid predicts the mirror frequency tuning. The planar property of the oscillator offers the possibility of a wafer-scale monolithically integrated source. Thousands of active solid-state devices can potentially be integrated in a high-power source for microwave or millimeter-wave applications. >

A grid amplifier

A 50-MESFET grid amplifier is reported that has a gain of 11 dB at 3.3 GHz. The grid isolates the input from the output by using vertical polarization for the input beam and horizontal polarization for the transmitted output beam. The grid unit cell is a two-MESFET differential amplifier. A simple calibration procedure allows the gain to be calculated from a relative power measurement. This grid is a hybrid circuit, but the structure is suitable for fabrication as a monolithic wafer-scale integrated circuit, particularly at millimeter wavelengths. >

Wave techniques for noise modeling and measurement

The noise wave approach is applied to analysis, modeling, and measurement applications. Methods are presented for the calculation of component and network noise wave correlation matrices. Embedding calculations, relations to two-port figures-of-merit, and transformations to traditional representations are discussed. Simple expressions are derived for MESFET and HEMT noise wave parameters based on a linear equivalent circuit. A noise wave measurement technique is presented and experimentally compared with the conventional method. >

Bow-tie antennas on a dielectric half-space: Theory and experiment

A new formulation is discussed for the rigorous calculation of the radiation pattern of a bow-tie antenna of finite length and infinitesimal thickness, placed on a lossless dielectric substrate. The analysis is based on a representation of the current density on the metal surface of the antenna as a sum of an imposed (quasistatic) term and a set of current modes with unknown amplitudes. Free-space fields that are expressed in terms of continuous spectra of symmetrized plane waves are matched to the current modes using the method of moments. The resulting set of equations are solved for the unknown current amplitudes. The calculations show that for increasing bow length the antenna impedance spirals rapidly to a value predicted by transmission line theory. The theory also shows that the E -plane pattern of a two wavelength, 60\deg bow-tie antenna is dominated by low-loss current modes propagating at the dielectric wavenumber. As the bow tie narrows, the loss of the modes increases, and the dominant wavenumber tends to the quasistatic value. Pattern measurements made at 94 GHz are shown to agree well with theoretical predictions. Measurements for a long-wire antenna, a linear array of bow-tie elements, and a log-periodic antenna are also presented.

Monolithic millimeter-wave two-dimensional horn imaging arrays

A monolithic two-dimensional horn imaging array has been fabricated for millimeter wavelengths. In this configuration, a dipole is suspended in an etched pyramidal cavity on a 1- mu m silicon-oxynitride membrane. This approach leaves room for low-frequency connections and processing electronics. The theoretical pattern is calculated by approximating the horn structure by a cascade of rectangular-waveguide sections. The boundary conditions are matched at each of the waveguide sections and at the aperture of the horn. Patterns at 93 and 242 GHz agree well with theory. Horn aperture efficiencies of 44+or-4%, including mismatch and resistive losses, have been measured. A detailed breakdown of the losses is presented. The coupling efficiency to various f-number imaging systems is investigated, and a coupling efficiency of 24% for an f0.7 imaging system (including spillover, taper, mismatch and resistive losses) has been measured. Possible application areas include imaging arrays for remote sensing, plasma diagnostics, radiometry and superconducting tunnel-junction receivers for radio astronomy. >

Transmitter Architectures Based on Near-Field Direct Antenna Modulation

A near-field direct antenna modulation (NFDAM) technique is introduced, where the radiated far-field signal is modulated by time-varying changes in the antenna near-field electromagnetic (EM) boundary conditions. This enables the transmitter to send data in a direction-dependent fashion producing a secure communication link. Near-field direct antenna modulation (NFDAM) can be performed by using either switches or varactors. Two fully-integrated proof-of-concept NFDAM transmitters operating at 60 GHz using switches and varactors are demonstrated in silicon proving the feasibility of this approach.

Far-infrared microbolometer detectors

The bismuth microbolometer is a simple, easily made detector suitable for use throughout the far-infrared, which has been integrated with a variety of planar antennas. The general thermal properties of these devices and some of the constraints on bolometer materials are discussed. The fabrication and performance of several different types of microbolometers and microthermocouples are described.