Premium
Comparison between different formulations of the transverse resonance field‐matching technique for the three‐dimensional analysis of metal‐finned waveguide resonators
Author(s) -
Bornemann Jens
Publication year - 1991
Publication title -
international journal of numerical modelling: electronic networks, devices and fields
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.249
H-Index - 30
eISSN - 1099-1204
pISSN - 0894-3370
DOI - 10.1002/jnm.1660040106
Subject(s) - resonator , waveguide , computation , transverse plane , modal , matching (statistics) , boundary value problem , field (mathematics) , resonance (particle physics) , boundary (topology) , interface (matter) , matrix (chemical analysis) , eigenmode expansion , compiler , scattering , mathematical analysis , physics , acoustics , computer science , optics , mathematics , algorithm , materials science , vibration , mechanics , engineering , normal mode , structural engineering , particle physics , maximum bubble pressure method , composite material , bubble , programming language , statistics , polymer chemistry , pure mathematics
Two different formulations of the transverse resonance field‐matching technique are applied to evaluate the modal spectrum of metal‐finned waveguide structures. In the theoretical treatment, Method 1 uses boundary conditions prior to interface relations and allows the number of expansion terms to be selected according to the dimensions of the cross‐section subregions. Method 2 imposes boundary and interface conditions in reversed order but requires the same number of expansion terms to be used in various subregions. The two procedures are then compared with respect to their influence on the modal scattering matrix computation of metal‐finned waveguide resonators. Method 2 shows excellent agreement with measurements, but it is restricted to configurations with relatively thin fins and moderate slot widths. Although Method 1 may be applied to more general structures, this procedure requires a higher computational effort and leads to slightly different results. However, it constitutes a powerful PC‐operational alternative whenever an extended precision compiler, as required for Method 2, is not available.