Larry D. Bacon

Evaluation of a thin-slot formalism for finite-difference time-domain electromagnetics codes

A thin-slot formalism proposed by Gilbert and Holland (1981) for use with finite-difference time-domain (FDTD) electromagnetics codes has been evaluated in both two and three dimensions. This formalism allows narrow slots to be modeled in the wall of a scatterer without reducing the spatial grid size to the slot width. In two dimensions, the evaluation involves the calculation of the total fields near two infinitesimally thin coplanar strips separated by a gap. A method-of-moments (MoM) solution of the same problem serves as a benchmark for comparison. Results in two dimensions show that up to 10 percent error can be expected in total electric and magnetic fields both near ( lambda /40) and far (1 lambda ) from the gap. In three dimensions, the evaluation is similar. The finite-length slot is placed in a finite plate and an MoM surface patch solution is used for the benchmark. These data show that slightly larger errors can be expected. >

On the limitation of the weak-coupling assumption for crosstalk analysis based on quasi-TEM propagation

The limitation of the weak-coupling assumption for the case of two transmission lines with a common ground is examined. The analysis is based on the mathematical structure of solutions to the telegraphers equations, and, therefore, quasi-transverse electromagnetic (TEM) propagation is assumed. All loss is neglected, and the lines are assumed to be symmetric. Requirements that are necessary for the weak-coupling assumption to hold in a homogeneous medium as well as in a medium that possesses certain types of inhomogeneities are derived by examining both the fully coupled and the weakly coupled forms of the telegraphers equations. It is shown that terminating impedances influence the validity of the weak-coupling solution locally to resonances but have little effect elsewhere. Theoretical and experimental results are given in both the frequency and time domains for several geometries. A solution to the weak-coupling form of the telegraphers equations is provided. >

Real-time RF spectrum analyzer: Components and system development

Many wireless communication links quickly hop between narrow frequency channels. Many such connections can occur simultaneously in the same band and hop in a standard pseudorandom frequency pattern dwelling a predetermined time in each channel. To sense modern communication, a real-time spectrum analyzer is very useful. One large advantage of real time analysis is that it only records data in active channels because it can determine the activity in each time interval. Another advantage is that communication that is not adhering to FCC standards can also be discerned. A van full of RF amplifiers, digitizers, and Fourier analysis equipment has been used for this job in non-real time. We chose to attack this design by providing many identical signal paths, one for each frequency channel. A number of RF components had to be developed to make this circuit power efficient and fit a small footprint. The initial broad band signal from an antenna sees low noise amplification (LNA) and then is divided into many identical RF signal paths using silicon germanium integrated circuits (SiGe RFICs). Each of these RF signals is filtered by one filter in a ladder of frequency adjacent SAW filters. The output of each SAW device is compared with the RF power seen in the previous time interval to see if there is less or more of a signal. Up to this point the system has been low- power analog. Once the RF channel powers are quantified, the system uses a digital signal processor (DSP) to further analyze signal characteristics.

A Robust Approach for the Analysis of EMI/EMC Problems With Nonlinear Circuit Loads

The analysis of electromagnetic coupling in nonlinear circuit simulations requires a bidirectional, fully consistent approach. Nonlinear responses of semiconductor devices in electronic circuit components can change the impedances seen at circuit nodes, changing the boundary conditions encountered by impressed electromagnetic fields and, thus, changing the characteristics of the energy coupled from these external fields into that circuit. It is important to include the coupling in the circuit simulation self-consistently because this allows us to accurately predict the responses to various EMI/EMC problems of interest. It is also important to predict circuit responses efficiently because that opens the door to statistical applications for the technique being used. In this paper, we review a technique that we have developed called A Thevenin Equivalent Network Approach. This approach is shown to be quite robust in that it is computationally efficient; it can be implemented in a variety of commonly available circuit solving codes; it already includes a few additional techniques required to enhance its implementation in those codes; and it is quite accurate.

Device Technology Investigation: Subsystems Packaging Study: Feasibility of PCSS - Based Pulser for Highly Portable Platforms

This report summarizes an investigation of the use of high-gain Photo-Conductive Semiconductor Switch (PCSS) technology for a deployable impulse source. This includes a discussion of viability, packaging, and antennas. High gain GaAs PCSS-based designs offer potential advantages in terms of compactness, repetition rate, and cost.

Theory and simulation of a high-power microwave convertor for current-density-modulated, annular, relativistic electron beams

The efficiency of converting the kinetic energy of a current‐density‐modulated, annular, relativistic electron beam into high‐power microwaves depends on details of the beam, and the convertor geometry. The latter issue is investigated in this paper for a particular choice of convertor. Maxwell’s equations are solved for the case in which a structure comprised of a cylindrical cavity coupled to a waveguide by an impedance transformer is excited by a symmetric, transverse magnetic wave. The field solutions are employed to determine the effect that the transformer/waveguide system has on the resonant properties of the cavity. An example is given in which the analytic model is used to obtain maximum power output in a numerical simulation.

Resonances in electromagnetic penetration of deep slots

The author discusses the influence of depth on the transmission of narrow slots and presents curves of calculated and measured transmission that exhibit depth resonances. The measurements are made by illuminating the front face of an aluminum box that is 50 cm on each side. At these resonances, the transmission is shown to be much higher than predicted by zero-depth models. >

An Active Thevenin Equivalent Network Approach to EMI/EMC Problems.

Nonlinear responses of the semiconductor devices in electronic devices can change the impedances seen at the circuit nodes, changing the boundary conditions encountered by impressed electromagnetic fields and thus the field coupling. We have developed the Active THevenin Equivalent Network Approach (ATHENA), which allows us to include electromagnetic coupling in nonlinear circuit simulations in a bidirectional, fully consistent way. Including the coupling in the circuit simulation self-consistently is important because it allows us to predict responses to EMI/EMC both correctly and efficiently, opening the way to predicting response statistics.

Loop-to-loop coupling.

This report estimates inductively-coupled energy to a low-impedance load in a loop-to-loop arrangement. Both analytical models and full-wave numerical simulations are used and the resulting fields, coupled powers and energies are compared. The energies are simply estimated from the coupled powers through approximations to the energy theorem. The transmitter loop is taken to be either a circular geometry or a rectangular-loop (stripline-type) geometry that was used in an experimental setup. Simple magnetic field models are constructed and used to estimate the mutual inductance to the receiving loop, which is taken to be circular with one or several turns. Circuit elements are estimated and used to determine the coupled current and power (an equivalent antenna picture is also given). These results are compared to an electromagnetic simulation of the transmitter geometry. Simple approximate relations are also given to estimate coupled energy from the power. The effect of additional loads in the form of attached leads, forming transmission lines, are considered. The results are summarized in a set of susceptibility-type curves. Finally, we also consider drives to the cables themselves and the resulting common-to-differential mode currents in the load.

Measurement of the dielectric properties of dispersive materials over a wide frequency range

The propagation of electromagnetic waves through dispersive media forms the basis for a wide variety of applications. Rapid advances in materials have produced new products with tailored responses across frequency bands. Many of these new materials, such as radar absorbing material and photonic crystals, are dispersive in nature. This, in turn, has opened up the possibility for the exploitation of these dispersive dielectric properties for a variety of applications. Thus, it is desirable to know the electromagnetic properties of both man-made and natural materials across a wide frequency range. With the advent of transient pulsers with sub-nanosecond risetimes and rates of voltage rise approaching 10**16 V/s, the frequencies of interest in the transient response extend to approximately the 2 GHz range.. Although a network analyzer can provide either frequency- or time- domain data (by inverse transform), common TEM cells are only rated to 0.5 to 1.5 GHz - significantly below the maximum frequency of interest. To extend the frequency range to include 2 GHz, a TEM cell was characterized and a deembedding algorithm was applied to account, in part, for the limitations of the cell. The de-embedding technique is described along with such measurement issues such as clear time and sneak around. Measurements of complex permittivity of common drinking water are shown. This frequency extension will lead to more expansive testing of dielectric materials of interest.