2956 IEEE TRANSACTIONS ON MICROWAVE THEORY AND TECHNIQUES, VOL. 61, NO. 8, AUGUST 2013 Precise Measurement Using Coaxial-to-Microstrip Transition Through Radiation Suppression Ondrej Moravek, Student Member, IEEE, Karel Hoffmann, Senior Member, IEEE, Milan Polivka, Member, IEEE, and Lukas Jelinek, Member, IEEE Abstract—This paper presents a study about the radiation problem of coaxial-to-microstrip launchers and suggests their improvement with a novel design. The proposed solution is based on a coaxial-to-microstrip transition enclosed in a parallel-plate transmission line that has its cutoff frequency above the working frequency band of interest. Any radiated field is quickly attenu- ated because it is propagating inside a subcritical parallel-plate transmission line. The proposed method is extensively analyzed in the CST Microwave Studio, and simulation results are verified on fabricated test-fixture by multiple measurements. The proposed solution improves the accuracy and reduces the uncertainty during measurements on a microstrip. Index Terms—Calibration, measurement techniques, microwave measurements, radiation effects, vector network analysis (VNA). I. INTRODUCTION A LONG with advances in microwaves, there was a great need for feeding the microwave signal from coaxial to planar transmission lines. A large number of interesting and in- novative designs were proposed since the 1980s [1]–[6] that were suitable either for microstrip or coplanar waveguide trans- mission lines. Some of these concepts intentionally shape elec- tromagnetic (EM) field distribution at the transition to improve microwave performance and to reduce the effect of the discon- tinuity. The quality of such a transition was assessed mostly by a magnitude of a reflection coefficient and by its mechanical us- ability. Now, it turns out that the concept of these transitions is even more complicated. Radiation of an EM wave originating at the coaxial-to-mi- crostrip transition is a recently discovered issue [7] that influ- ences precise microwave measurements in open transmission lines inthe -band and above. Considerable attention has been given to the radiation of the leaky-wave modes from the planar transmission lines [8], [9]. However, the radiation mode at the transition should not be confused with the well-described leaky- Manuscript received May 08, 2013; revised June 15, 2013; accepted June 19, 2013. Date of publication July 23, 2013; date of current version August 02, 2013. This work was supported in part by the Student Grant Competition (SGS) Pro- gram SGS10/271/OHK3/3T/13 of Czech Technical University in Prague spon- sored by the Ministry of Education, Youth and Sports of the Czech Republic and the EMRP Project ”SIB62 Metrology for new electrical measurement quantities in high-frequency circuits,” jointly funded by the EMRP participating countries within EURAMET and the European Union. The authors are with the Department of Electromagnetic Field, Faculty of Electrical Engineering, Czech Technical University, Prague, Czech Republic (e-mail: ondrej.moravek@fel.cvut.cz). Color versions of one or more of the figures in this paper are available online at http://ieeexplore.ieee.org. Digital Object Identifier 10.1109/TMTT.2013.2272380 wave mode. Some interesting studies on this topic have been al- ready published. For example, it has been shown that the radia- tion clearly disturbs commonly used calibration methods on mi- crostrip [7] and introduces other interesting phenomenons [10] associated with the radiation during measurement with vector network analyzer (VNA). The fact that a radiated wave inter- feres [7] with the quasi-TEM (QTEM) mode on microstrip helps to explain why the most frequently used calibration methods are affected by the radiation. It was also observed and verified [11], [12] that the magnitude of radiated power is proportional to the phase of the reflected wave, and, thus, it depends directly on the distance of the dis- continuity from the transition. This means that the error model is dependent on the device under test (DUT), which disrupts the standard calibration procedures. A condition of a consistent and linear error model which has to stay constant for each measured DUT applies for all calibra- tion methods suitable for microstrip [13]–[18]. However, this condition is not satisfied in the cases where the radiation intro- duces a multimode propagation where both modes (QTEM and radiated wave) interfere with each other. Different hardware approaches supported with three-dimen- sional (3-D) EM full-wave simulations were proposed [19], [20] to eliminate the radiation problem. The first paper [19] suggests modifications in the SMA coaxial-to-microstrip transitions. It is based on enclosing the transition into a conducting metal ring. The other solution suggests placing the microstrip line between two conducting sidewalls from both sides in such distance so it will not influence the microstrip QTEM mode. This unfor- tunately introduces possible waveguide-like behavior of such structure. Some higher order (waveguide) modes would have to be dealt with, and additionally, it is not suitable for all possible dimensions of DUT. The purpose of this paper is to present a physical explanation of the radiation from the transition and to show another possible design of coaxial-to-microstrip transition that would sufficiently suppress the unwanted radiated wave without introducing any disadvantages in usability. Experimental and simulation results of the proposed structure are given, and it is shown that the published hardware solution improves the measurement accu- racy when using the common calibration method (1-port and 2-port Short-Open-Load-Thru (SOLT) [16]). Unlike the previ- ously published work [20], this paper introduces versatile con- figuration of the test-fixture without any practical limitations. Authors would like to note that the presented radiation problem should not be confused with leaky-wave modes on microstrip, which would not be present due to dimensions of the microstrip (i.e., substrate thickness) used throughout this paper. Finally, 0018-9480/$31.00 © 2013 IEEE