Author: Anil Pandey
An increasing number of RF and Microwave applications at the sub-millimeter and millimeter frequency ranges in different fields such as communications, imaging, automotive radar and tracking radar has increased the demand for the components and systems in this range. To handle the high-power requirement in these frequency ranges, waveguide-based components are used as a power feed network. In the mmWave, the waveguide is normally adopted as the main transmission line instead of a coaxial cable due to easier manufacturability and lower loss. Components based on planar technology like monolithic microwave integrated circuits (MMICs) are relevant. Therefore, a low insertion loss and broadband transition between the microstrip line and waveguide is very crucial for these applications. In addition, at millimeter and terahertz frequencies, most measuring equipment is waveguide-based due to the lack of connectors that’s why the transition is needed. There are several types for microstrip-to-waveguide transitions based on the application like the design based on the ridged waveguide, antipodal finline, coupling probe, dipole antenna, substrate integrated waveguides (SIW), iris coupling, etc.
Design-1: E-Plane Probe Transition with Back Short
Among these, the E-plane probe coupling method is very frequently used. In this transition, a microstrip line is inserted through an aperture in the E-plane of a rectangular waveguide that couples TE10 mode of the waveguide to the quasi-TEM mode of the microstrip line. In the E-plane probe transition network, the input port is perpendicular to the output port, sometimes this makes the realization of devices or antennas very difficult in limited space as addition bends are required. This transition consists of a printed probe on one side of the dielectric substrate inserted into rectangular waveguide with back short. This type of transition is very sensitive to the position of the dielectric substrate and microstrip line with respect to the waveguide. However, there are some drawbacks to this type of transition. The channel width and height, that holds the microstrip substrate and the E-plane probe, should be very small so that all the waveguide modes are under cut-off and only the microstrip quasi-TEM mode can propagate within the band of operation.
Below is the design of a waveguide-to-microstrip transition based on the E-plane probe, designed in mmWave (65-85 GHz) for automotive radar applications. A coupling probe at one end of the microstrip line is introduced into the perpendicular waveguide through an aperture in the waveguide broad wall. The copper probe is designed as a rectangular sheet. One end of waveguide is short-circuited by the metallic plate. Liquid Crystalline Polymer (LCP) is used as the substrate material for reducing the transition loss. The thickness of the substrate is set as 0.79mm. The microstrip line characteristic impedance is 50Ω. The distance between the probe center and the waveguide back-short is a quarter of the WR12 waveguide wavelength at 75 GHz. The aperture size is optimized to guide the wave coupled to the microstrip line effectively and to minimize its effect on the field distribution in the waveguide.
Lateral radiation introduce leakage that is suppressed by using stitching vias to connect both side of ground planes on the substrate with the upper and lower waveguides. Also, in order to reduce the leakage from the waveguide opening window, the width of the opening should be smaller than the width of the cut-off condition. A radial stub has been used at the end of the probe to match the impedance of the network for broadband performance.
S-parameters of the return loss S11 and the transmission loss S21 are calculated by a full-wave Finite Element Method (FEM) based electromagnetic simulator. The simulation result of the waveguide-to-microstrip transition is shown in Fig. 3. The return loss is less than -20 dB within 62 GHz – 78 GHz. The maximum return loss is -22.8 dB at 73 GHz. The insertion loss is greater than -0.21 dB within 62-78 GHz. The minimum insertion loss is -0.1 dB at 73 GHz. The bandwidth of the transition is about 22% for less than -20 dB return loss.
The electric current on the metallic probe couples to the magnetic field of TE10 dominant mode of the rectangular waveguide as shown in below picture. Via holes reduces the leakage of parallel plate mode transmitting into the substrate. Impedance matching is achieved by controlling the length of the probe and the distance of the back short waveguide.
Design-2: Half-Height Waveguide to Microstrip Line Transition
To reduce the transition dimension and simply the design, a half-height rectangular waveguide to microstrip line transition network is used. The gradual transition from the microstrip line to the waveguide is used to transfer impedance smoothly. The reflection coefficient at the feed point depends on the probe length, a distance of the waveguide short and probe width. To make sure the efficient electric coupling, the probe is always located at the maximum electric field in the structure. The below figure shows the electromagnetic simulation model of half-height waveguide WR12 to 50Ω microstrip line. The planar circuit is supported on a quartz substrate or 10 mils. The copper probe patch is perpendicularly inserted into the center of the broad wall. The plane is perpendicular to the transverse section and parallel to the E-plane of the half-height waveguide. It can be noted that in design-1 discussed above, Microstrip probe is perpendicular to E-plane.
The simulation result of the back-to-back design is shown in the below figure. Based on FEM solver results, it is found that transition structure has S11 better than -20 dB from 70 GHz to 80GHz. S12 is less than 0.12 dB over the operating frequency band.
Design-3 Oversized Waveguide to microstrip line transition for Radial Power combiner/ divider
Power amplifiers (PA) are found in several millimeter-wave systems. Radar and satellite communications are two systems where high-power is very important. MmWave communication systems require high transmitter power because atmospheric attenuation at mmWave (60GHz) frequency is quite high. The power needs at this frequency are up to several watts and a single Solid-State Power Amplifier (SSPA) cannot meet this requirement. Radial high-power combing technology overcomes this design challenge. Radial divider/combiner distributes the signal of one common input port to the N-output waveguides cavity. This power is coupled with a cavity microstrip line resonator and supplied to SSPA. The output of SSPAs is again supplied to N-Waveguides that are combined to a single output port as shown in the figure. The radial divider/combiner does not use corporate or cascaded structures and, thus, routing losses are minimized, and overall insertion losses of the whole structure are very low.
Axially symmetric oversized coaxial waveguide power combiner utilizes a planar microstrip probe to transform TE10 mode to quasi TEM mode. This transition network has broadband with high-efficiency performance, which has been used widely in many mmWave systems. To provide a smooth impedance match from the input port to the oversized coaxial waveguide, a nonuniform coaxial taper transition circuit is used. The inner conductor is inserted into the center of the gradual taper in order to provide a perfect electrical connection. The outer and inner conductors of the coaxial taper transition have been tapered at two different planes rather than on the same plane, which can provide a good impedance match.
The simulated return loss and insertion loss plot of the planar probe combiner are shown in the below figure. Return loss (S11) is better than -10dB over the band (62 GHz to 87 GHz) giving wideband (35% bandwidth) performance. The isolation between the peripheral ports is better than 12 dB. This type of radial power combiner/divider establishes the advantages of easy design and fabrication, perfect compatibility with the mmWave integrated circuitry, low insertion loss and high power-combining efficiency.
Design-4 In-Line Stepped Transformation H-plane Waveguide to Microstrip line Transition
The TE10 rectangular waveguide mode can be coupled to quasi-TEM of microstrip mode by using inline configuration. In this design, the field propagation direction in the microstrip line is the same as the waveguide. A wideband impedance transformer using stepped ridge waveguide provides excellent transition performance. Ridge waveguide has a smaller equivalent impedance which is easier to match with the coaxial and microstrip line. Waveguide to Microstrip line transition is designed using a multi-section stepped transformer as shown in Figure. In this design, the inner coaxial conductor extends across the interior of the waveguide to terminate on a metallic probe to assist in impedance tapering. An in-line waveguide-to-microstrip transition design was proposed by Yao  in 1994.
The probe is built using a microstrip line matched to the waveguide through a ridged configuration, on which the microstrip line is attached as shown in the below Figure. A Chebyshev 4-section impedance transformer provides the matching from the stepped transition to the full waveguide. The probe lies on a plane parallel to the wide side of the waveguide and it is inserted from the back-short of the waveguide such that the E-planes of the two devices are parallel. The microstrip line is centered along the wide side.
The design is simulated using a FEM solver. A return loss greater than -15dB over the entire operating band (71 GHz to 79 GHz) is obtained for back-to-back configuration as shown in the below picture. The insertion loss is lower than 0.2, so a good power transmission is provided.
Design-5 Inline Waveguide to Microstrip line transition using a Slope-Probe
Fabrication of stepped ridge probe, discussed in design-4, is difficult, and dimensions are sensitive to result. Instead of the stepped transformer, a smooth copper transition as shown below in the picture can be used to match waveguide impedance to microstrip line impedance. In this way, the fundamental TE10 mode of the rectangular waveguide is converted progressively to the quasi-TEM microstrip mode in a broadband inline configuration. The microstrip line is printed on a dielectric substrate and the end of the substrate along with the cavity enclosure is narrowed to concentrate the fields near the metallic strip. The short portion of the microstrip line is inserted into the waveguide and placed seamlessly underneath the probe.
Design-6: Inline Waveguide to Microstrip Line Transition through a Coaxial Line
Wheeler  first investigated In-line waveguide transitions in the 1950s, when he studied the transition from waveguide to a 50 ohms coaxial line. Waveguide to Microstrip line (through SMA connector line output port) transition is designed using a multi-section stepped transformer as shown in Figure. In this design, the inner coaxial conductor extends across the interior of the waveguide to terminate on a metallic probe to assist in impedance tapering.
Design is based on two key factors; the first factor is mode conversion from the dominant mode of the input transmission medium to that of output transmission medium and impedance matching is the second facto. In rectangular waveguide to coaxial line transition, TE10 mode in the waveguide is converted to the TEM mode of the coaxial line. The characteristic impedance of rectangular waveguide is a function of frequency. Therefore, to match the rectangular waveguide impedance to the 50 Ω coaxial line impedance over the desired frequency range, a multi-step matching transformer is needed. Simulated return and transmission loss are shown below in the figure.
In some applications, microstrip antennas are placed on the top surface of design and are fed with an in-line waveguide as shown below in the figure. One advantage of such design is that it can be easily integrated with the Microstrip feed-line of antenna or can be directly connected to patch antenna.
Design-7 E-Plane Waveguide to Microstrip line transition through Coaxial line
In the E-plane transition referred to as orthogonal transition, the center conductor of the probe (coaxial line) is inserted into the waveguide. This excites the TE01 mode in the waveguide. The probe is placed in such a way that the reflected EM fields get combined in phase with the incident EM fields. The distance between the back wall and probe center is optimized to achieve good impedance matching. Typically, it is kept at λg/4. E-plane probe is the most popular type of transition but it gives narrow band performance unless additional matching elements are included which increases the physical size. The coaxial-line and probe are perpendicular to microstrip and waveguide. The probe is inserted to the waveguide at the center of the waveguide wide side. S11 result is shown in the figure that is better than -10 dB in all the band of interest.
 A. Pandey, Practical Microstrip and Printed Antenna Design, Artech House, Norwood, MA, 2019.
 Wheeler, G. Broadband waveguide-to-coax transitions. In Proceedings of the 1958 IRE International Convention Record, New York, NY, USA, 21–25 March 1966; Volume 5, pp. 182–185.
 Yao, H.-W.; Abdelmonem, A.; Liang, J.-F.; Zaki, K.A. Analysis and design of Microstrip-to-Waveguide Transitions. IEEE Trans. Microw. Theory Tech. 1994, 42, 2371–2380.
 A. K. Pandey, “Design of a cosecant square-shaped beam pattern SAR antenna array fed with square coaxial feeder network,” 2013 European Microwave Conference, Nuremberg, 2013, pp. 1699-1702.