Calculation of errors in the finite difference time domain method when modeling thin wires separated by a single cell of the computational grid

Abstract

The finite difference time domain method contains additional techniques for modeling thin wires whose diameter is substantially smaller than the cell size. Among them, there is a group of techniques where thin wires are modeled by correcting the medium surrounding the wire. Although in this case only the wires located along the electric field nodes can be modeled, techniques of this group do not cause calculation errors if the wires are located in the absorbing boundary conditions necessary to model thin wires with infinite length. Several techniques have been compared for the case when two wires are separated by one-cell gap. Influence of magnetic field modification around the gap between wires on calculation results has been examined. By calculations with the dipole antenna, it has been shown that correction of the magnetic field around the gap does not increase calculation accuracy (at least for the considered cases). However, this correction eliminates numerical instability. In this paper, it has been proposed to apply a magnetic field correction at the ends of the conductor only in cases where the conductors are separated by one cell, but not at each end of the conductors, as previously suggested.

Keywords

• finite difference time domain (FDTD) method
• numerical instability
• dipole antenna
• wire modeling technique
• calculation errors