S.J. Kubina

Fields of complex surfaces using wire grid modelling

In the resonance range of frequencies, the scattered field from a complex surface such as a ship or aircraft depends on the geometry of the scatterer as a whole rather than on individual scattering centers. Excellent results are obtained by replacing the complex surface with an equivalent grid of wires or wire grid model. The calculation of the currents flowing on the wires of a grid by a moment method solution of an electric field integral equation is summarized. Assumptions made in the formulation lead to modelling guidelines which are local restrictions to the geometry of the wires of the grid. A set of wire grid guidelines or principles to aid in the global design of wire grid models is presented. The objective is an orthogonal grid of square cells of uniform side length. The wire grid guidelines provide a basis for compromise in grid design. They help identify weaknesses in wire grids. These principles are applied to model an ellipsoid and a sphere, with good agreement with measured data, and the exact solution, respectively. >

Power line tower models above 1000 kHz in the standard broadcast band

It is shown that the resonant behavior of power lines at higher frequencies is quite dependent on the geometry of the crossarms that the tower uses to carry the phase wires. Tower models (including crossarm detail) which can reproduce the Thornhill measurements at 1605 kHz and the scale-model measurements from 1000 to 1900 kHz are developed. One such tower model is simple enough to allow the analysis of long runs of power line. Such models are useful in estimating scattering from power lines across the entire AM broadcast band.<<ETX>>

Stub detuners for free-standing towers

A standard broadcast AM radio stations antenna pattern has minima in the directions of other cities with stations on the same frequency. However, scattering can raise the field strength in a certain direction to unacceptably high levels. The scatterer must then be identified and modified or detuned so that it no longer scatters the field. Several detuner designs are presented for installation on a powerline tower for the suppression of the scattered field. The emphasis is on the assessment of the bandwidth over which scattering is suppressed, and on how the detuner can be modified to increase the bandwidth . The performance of detuners is studied by computing the scattered field from a type V1S power line tower using the Numerical Electromagnetics Code. The uses of a detuning skirt or detuning stub are also investigated. It is shown that the bandwidth can be improved by increasing the stub length, as in the elbow stub design. It is further shown that a large increase in bandwidth can be achieved by separating the stub significantly from the tower, thus increasing the energy loss in the resonant system. The effectiveness of the proposed stub designs is investigated by the measurement of the base impedance of the scale model tower detuned with each stub design. The measured impedances bear out the conclusions that the bandwidth of a capacitively tuned stub is improved by increasing the stub length, that the bandwidth of a stub can be increased by separating the stub wire from the tower body, and that the wire stub designs offer as good or better performance than do the capacitively tuned stub designs. >

Validation of FDTD RCS computations for PEC targets

The radar cross-section (RCS) of perfectly conducting (PEC) targets calculated by the finite-difference time-domain (FDTD) method is compared with measured data. For face-on incidence on a cube, excellent agreement is obtained over a 30-to-1 frequency range. For end-on incidence on a 200*20*20 cell rod, very good agreement is obtained over the measurement range of size from 0.5 lambda to 10 lambda rod length, provided 20 cells of free-space separate the rod from the absorbing boundary. For a 110*11 cell circular cylinder, very good agreement is obtained out to lengths of 10 lambda with 22 cells of white space. Thinner white spaces greatly reduce the computation time, but compromise the agreement with the measurement.<<ETX>>

Ground loss effects in power line reradiation at standard broadcast frequencies

Reradiation of a standard broadcast antennas signal from a high-voltage power line is often analyzed by computer modeling. The resonant behavior of a power line operating over lossy ground is investigated using the numerical electromagnetics code (NEC) which uses the Sommerfeld-Norton (SN) ground model, for a highly accurate computation of the interaction of a power line with lossy ground. It is shown that lossy ground damps the resonant response of the power line and so, substantially reduces the reradiated field. Since the SN ground option is costly in CPU time, the effects of lossy ground were approximated using the tower footing impedance concept. A systematic evaluation of the accuracy of the approximation is made. An approximate technique for incorporating the approximate techniques presented are sufficiently accurate that the economical method-of-images solution can be used in most cases. >

Scattering from power lines with the skywire insulated from the towers

This paper studies the distortion of a commercial AM radio stations radiation pattern due to scattering from a power line with steel lattice towers which carry a skywire that is insulated from all the towers. Despite the insulated skywire, the towers couple strongly to the skywire. The resonant behavior of the power line spans is similar to that described in previous papers with the skywire connected to all the towers. However, the resonant frequencies are much higher when the skywire is insulated and the strongest resonance falls above 1000 kHz for typical power line dimensions. This paper models a real site consisting of a directional broadcast array at 1380 kHz and a power line with 29 towers. Although the resonances of the power line fall at much lower frequencies than 1380 kHz, the antenna induces substantial RF current flow on the towers and skywires of the power line. The power line scatters strongly into the minimum in the stations radiation pattern and the stations protection requirements are seriously violated. Thus, the towers of the power line will need to be treated with detuning stubs to restore the pattern to its intended values. >

Validation of FDTD handset and head patterns by measurement

This paper compares the principal plane radiation patterns of a portable radio handset at 850 MHz computed with the finite difference time domain method with measured radiation patterns. The agreement for the handset in isolation is excellent. The computed patterns of the handset operating near a box model of the head are compared with measured patterns with very good agreement. Finally the computed patterns of the handset operating near a sphere model of the head are compared with measured patterns, with very good agreement, but not quite as good as for the box. The paper demonstrate the effect of simple head models on the radiation patterns of a vertical handset, and provides a validation of the FDTD computations.

Near fields of a portable radio in front of and on the far side of a model of the head

A portable radio such as a cellular telephone that is held against the side of the head induces fields in and around the head. This paper investigates the electric field strength in front of and on the far side of the head due to a vertically-oriented portable radio. The head is represented either by a box or by a sphere. The finite-difference time-domain method using a perfectly-matched layer absorbing boundary is used to compute the fields. The fields are measured using a three-axis probe moved over a plane by a robot positioning system. Good agreement is demonstrated between the measurements and the computations for field components strong enough for an adequate signal to noise ratio in the measured data.

Portable radio handset patterns in the presence of a model of the head

This paper examines the principal plane radiation patterns of a portable radio handset at 850 MHz operating near a model of the human head. The handset and head are solved by the finite-difference time-domain method. To validate the computed patterns, they are compared with measured radiation patterns, with excellent agreement for the handset alone and very good agreement for both the handset near the box model of the head, and the handset near the sphere head model. FDTD is then used to find the principal plane patterns of the handset near a realistic model of the head based on anatomical cross-sections. The principal plane patterns of the handset and anatomical head are shown to be very similar to those of the handset near a sphere of size comparable to the head.

Nonuniformity in a class of UTD curved surface fields near smooth, convex surfaces

Two different creeping wave formulations are examined in this paper. One is valid for source and field points on or near the curved surface, the other is valid for source and field points off the curved surface. This paper uses a numerical experiment with a circular cylinder to determine at which height the two different formulations overlap.