When designing slotted waveguide antenna, the most popular technique to feed the antenna is the use of a coax-waveguide transition.
A transition structure is required due to:
- Difference of guided modes: In the coaxial cable, the mode is TEM where both magnetic and electric fields are transverse (orthogonal to the propagation axe). Meanwhile in the waveguide, the mode is TE10.
- Difference of characteristic impedances: impedance in the coaxial cable is 50 Ohm and that of waveguide can be several times higher (depending on the structure), in the range of 300 – 400 Ohms. What does it mean “characteristic impedance”? Well, long story short, the geometry and the material of a transmission line defines a factor between the electric field strength and the magnetic field strength. That factor is “characteristic impedance”. It tells how the structure favor electric field compared to magnetic field.
This article addresses directly to a specific case: from N-type connector to WR975 waveguide but the procedure is generic and can be apply to other waveguide dimensions. The simulation setup is quite simple: we need a circular waveport at N-type connector and a rectangular waveport for the waveguide. Ansys HFSS is used to simulate the structure but CST and Feko or Comsol can do the job too. A distance slightly longer than the guided wavelength is chosen to make sure that the propagation is taken into account. The transmission coefficient S21 represents directly how well the energy is transferred from the coaxial cable to the waveguide. A “good” transition is the one where the transmission is close to 0 dB (means all RF energy is transferred through the transition and there is very small reflection) on the largest bandwidth that covers the frequency of interest.
How it works
The figure 3 gives an insight of how it works in the point of view of “electromagnetic field” (or antenna if you will). A monopole structure is naturally fed by the coaxial cable and this create an electric field aligned with the monopole. This electric field naturally excites the TE10 mode that has the same E-field pattern: the field is maximum at center; its polarization is normal to the large wall of the waveguide.
To optimize the transmission coefficient, we should analyze the structure in the circuit point of view.
We can model the transition structure by a LC circuit that match 50 Ohms of coaxial cable to 330 Ohms of the waveguide. The capacitor in the circuit depends on the gap between the post is to the waveguide wall. The inductor value depends on distance l_shunt between the monopole and the short-circuit wall.
Fig. 5 a) optimization procedure, b) simulated return loss after optimization
A simple parametric simulation in HFSS give a result quite nice. We get -30 dB return loss (thus 99.9% energy pass through the transition structure) at the frequency of interest.
In this article, the classic way of designing a coaxial-waveguide transition has been presented. To understand the physic of the structure (how it works), it is important to apply both point of view: antenna point of view and circuit point of view. Antenna point of view helps understand the field pattern in the waveguide and how to excite the TE10 mode by using monopole to naturally create the pattern. Circuit point of view helps determine which parameter to optimized. The study has been conducted by using 3D EM simulator.