On the convergence of the input impedance

[AN-SOF Antenna Simulator Group]

Input impedance convergence has been a problem that older codes like NEC have not been able to solve. The root of the problem lies in the use of an approximate kernel instead of the exact one, since only thin-wire approximations or extended versions of the kernel were known at that time.

The kernel is the core of the integral equation that must be solved to obtain the current distribution and the precision with which it is calculated directly impacts the results.

The figure shows the input impedance of a half-wave dipole vs. the number of segments, N, obtained by AN-SOF with exact kernel and NEC with extended kernel (wire radius = 0.005, gap = 0.025 wavelengths).