A classical eigenvalue mode-spectrum analysis of waveguides with multi-ridged cross sections is presented and applied to the design of narrowband waveguide components in rectangular and circular waveguide technology. Modifications of the modes of the empty waveguide enclosures are used as expansion functions and lead to a classical, real and symmetric eigenvalue problem. A simple yet efficient constraint function is introduced to satisfy boundary conditions for TM modes. The number and locations of ridges positioned in a regular rectangular or circular waveguide enclosure is arbitrary. Measurements and comparisons with results from existing full-wave modeling tools and commercially available field solvers verify the correctness and flexibility of the approach. waveguides into computer-aided design procedures for narrowband waveguide components, their mode sequences and related expansion coefficients must be known.A popular approach to determine the modal distribution of irregular waveguides is to subdivide the cross section into subregions, e.g. [3][4][5][6][7][8][9][10][11][12][13]15], which leads to a singular-value problem requiring the system determinant or, alternatively, the smallest singular value to vanish [9]. However, this method has several disadvantages. Firstly, it requires a search algorithm to determine the mode spectrum, which is computationally expensive and lacks accuracy, especially if cut-off frequencies of modes are located very close together or are even identical within the smallest variation acceptable in the search. Secondly, if the cross-sectional dimensions are varied within an optimization run, the boundary conditions with respect to the subdivisions might change, thus requiring that the altered configuration be solved and coded separately. Thirdly, using different subdivision schemes, e.g. horizontal versus vertical, leads to slightly different results for an entire waveguide component containing such cross sections. This is demonstrated in [11]. It is therefore advisable to develop a method that refrains from dividing the entire cross section into subregions. This is accomplished by setting up a classical eigenvalue formulation, e.g. [19,[22][23][24], and selecting appropriate basis functions. The mode-matching-finite-element method, e.g. [25], is especially useful in this respect. One issue to consider, though, is whether or not the resulting eigenvalue equation requires complex arithmetic to be applied to asymmetric matrices. In this respect, the approach presented in [26][27][28][29] is advantageous since the resulting matrices are real and symmetric, and, in addition, the approach is easily combined with the three-dimensional mode-matching technique (MMT). However, the basis functions used in [26][27][28][29] are polynomials involving the computation of Gamma functions, which limits efficient code implementation. Therefore, only simple discontinuities have been presented so far [29] as they require only a limited number of eigenmodes to be considered in the irregular cross sectio...