Major error and sensitivity analysis for characterization of laminate dielectrics on PCB striplines

Traveling-wave technique [1] has been proven as an effective way to extract dielectric constant (DK) and dissipation factor (DF) of laminate dielectrics on PCB test boards with unknown cross-sectional geometries and conductor surface roughness profiles. This technique is called S3 and is based on measuring the full set of complex S-parameters of a test board with an appropriate “through-reflect-line” (TRL) calibration pattern, and then separating conductor loss from dielectric loss using “root-omega” procedure. Currently layouts of test boards with TRL calibration patterns operating up to ~ 50 GHz have been developed. Analysis of sensitivity to different factors and errors associated with measuring DK and DF are important issues, especially as frequency increases. In this paper, two main groups of errors in DK and DF extraction are considered. The first group includes repeatability errors, quantified for the tests performed on multiple test boards with the same dielectric taken from the same manufacturer and the same batch. The second group includes systematic, or reproducibility, errors of measurements conducted on a single test vehicle. Herein, an approach to quantify these two types of errors in extracting DK and DF is presented.