Geometry description

Geometry options

Dimension limits

Geometry guidelines

Geometry options

Canonical structure mesh - The Canonical structure mesh option is used to define structures covered with a wire mesh. Structure options include a box, pyramid and truncated pyramid.

Catenary wire - The Catenary wire option is used to define a wire between two points with the catenary shape of a hanging cable.

Convergence test - The Convergence test option is used to define a convergence test for the problem. The choice of the number of segments is critical to the validity of the computation. The accuracy is dependent on the number of segments. However, an increase in the number of segments also increases the memory and computational time requirements. A convergence test can be used to determine a "reasonable" number of segments. A convergence test consists of calculating the conductance and susceptance for the problem as the number of unknowns is increased. Since the conductance and susceptance values both converge as the number of segments are increased, the convergence test can be used to determine the accuracy that can be expected for a given segmentation density (i.e., the number of unknowns per wavelength of wire).

Dimensions, Environments, Coordinates - This option allows the user to change dimensions, environments and coordinates.

Geometry points - The Geometry points option is used to define the coordinates of the geometry points for defining Straight wires and Wire meshes.

Geometry points iteration - The Geometry points iteration option is used to define increments for the geometry points. For each geometry iteration, a separate file will be produced. Caution is advised, since the indiscriminate use of this option may create a large number of files. This can quickly consume all available disk space.

Helix/spiral - The Helix/spiral option is used to generate a helix, a log spiral, or an Archimedes spiral approximation using straight wires.

Numerical Green's Function - The Numerical Green's Function is used to define the last wire in the Numerical Green's Function. It is possible to improve the computational time for a given problem using the Numerical Green's Function. For example, consider the analysis of a small antenna near a large structure. A computation is made for the complete problem. If the small antenna is the only part of the problem that is to be moved, the computational description of the large structure can be saved in a file and used in the subsequent calculations as the small antenna is moved.

Reflection - Reflection forms geometries having planes of symmetry by reflection of the geometry in the coordinate planes. Any combination of reflection along the x, y, and z axes may be used. The reflections are done in reverse alphabetical order. The geometry is first reflected along the z-axis. The geometry is then reflected along the y-axis. Finally, the geometry is reflected along the x-axis.

Redundancy check - Redundancy check will review the geometry description and remove any redundant geometry nodes or wires. The criterion for redundant geometry nodes is geometry nodes within a wire radius of each other. The wire radius is the largest radius of all wires connected to a geometry node. The user can change this criterion by a multiple of the wire radius.

Straight wires - The Straight wires option is used to define the attributes for each wire: number of segments, radius, geometry points of each end, and end caps.

Structure scaling - Structure scaling scales all dimensions of a structure by a constant.  This includes the location of the geometry nodes and the radii of the wires.

Symmetry - The Symmetry option is used to define when symmetry is to be used in the calculation of currents.

Tapered structure - The Tapered structure dialog box is used to generate a tapered structure using straight wires.  The tapering can be in terms of the lengths of the wires defining the tapered structure and the radii of each wire in the structure.

Text file input - This option allows the user to input a geometry description from a user-developed text file.

Transformations - Transformations move existing wires by rotation and translation. Copies of wires can be generated.

Wire arc - The Wire arc option is used to generate a circular arc of straight wires.

Wire mesh - The Wire mesh option is used to cover a surface with a wire mesh.

Spiral wires - Spiral wires reorders wires along a principal axis. This can considerably reduce the solution time if an axis is chosen along the model's greatest length. This is a function of the matrix solution routine used in Expert MININEC. The effect of reordering will only be apparent for large problems.

Dimension limits

Expert MININEC Broadcast Professional is dimensioned for 10000 current unknowns and 4000 wires. Expert MININEC Classic is dimensioned for 1250 current unknowns and 500 wires. The specific large problem computational capability depends on the internal memory available in the computational platform. For a personnel computer with 8 Mbytes of internal memory, a problem with approximately 700 current unknowns can be accommodated. Problems with 2000 current unknowns have been run successfully on platforms with 20 Mbytes of internal memory. The current computation of Expert MININEC is configured to indicate when a problem is too large for the available memory. However, Windows may not always provide this indication, and the computation will simply stop. If the percentage indicator of the current computation is not progressing after some period of time, the problem is too large for the available memory.

Geometry guidelines

Geometry guidelines permit the user to modify the criteria used in definition evaluation. Method of moments solutions to antenna problems are at best approximations. There are four key factors that determine how close a computer analysis is to the "real world" performance of an antenna. These factors are

numerical methods employed and how well these methods are implemented in a computer code,

inherent accuracy of the computer,

how well the antenna numerical model corresponds to the physical model,

user's experience in recognizing problem areas.

The last factor requires the user to be fully aware of potential difficulties throughout the modeling process, from initial setup to output interpretation. User experience is gained through the process of code validation. The process of code validation is one of verifying the accuracy, and perhaps relevance, for a few well selected configurations or scenarios. This process will lead to "a warm and fuzzy feeling" that the code can provide reliable and accurate predictions. A perception of the expected accuracy within the stated limitations and scope of the program is achieved.

New users should always verify that the code produces modeling results that are self consistent. The user should also try reproducing the results of others, verifying those calculations with available canonical solutions or measurements, as suggested by the documentation and experience reported by others. Satisfactory results always contribute to the user perception of validity and self-confidence in using the code. Training and experience go hand in hand with the successful application of the electromagnetic modeling program. The training may be formal in a course or simply self-taught by extensive reading and trial and error.

A set of modeling guidelines have been developed to aid the user in constructing a numerical model of an antenna. These guidelines are in keeping with the assumptions of the Expert MININEC Series formulation. These are guidelines. For a specific problem the user may be able to obtain useful engineering answers outside these guidelines.

Individual wire guidelines

segment length (wavelengths): Each wire should be subdivided into segments comparable to less than 0.1 wavelengths. There is significant loss of accuracy if the segments are greater than 0.2 wavelengths. The segments can be made very small with respect to a wavelength. Good results are obtained to a lowest frequency of 3E-13 MHz. Such small segments do not work as well for loops.

segment length/radius ratio: The shortest permissible segment is usually determined by the wire radius. With the thin wire assumption, the ratio of segment length to wire radius should be maintained greater than 8. Some reasonable results can be obtained down to 2. For thin segments good results have been obtained for ratios of 1E7.

radius (wavelengths): The wires must be "thin" because the current is assumed to flow axially on the wire with no circumferential component. The wires are considered to fully satisfy the guidelines if they are thinner than .01 wavelengths. If wires are fatter than .03 wavelengths, the results may be questionable.

Wire junction guidelines

segment length ratio: Conservatively, wires making up a junction should not be too dissimilar in their segment lengths at the junction. For best results the ratio should be less than 2. Differences greater than 5 should be avoided.

radius ratio: For best results the ratio of the radii of wires at a junction should be less than 10. The Expert MININEC Series formulation has shown reasonable results to ratios of 100.

Crossed wires guidelines

number of wire radii: This defines the distance from the center of the wire in number of radii that is used to determine when wires cross. Crossed wires that touch will not be computed correctly, because the junction is not modeled unless the intercept point is a geometry point. In addition very close wires should be avoided.