Measurement-Based Modeling of PCB-to-Coaxial Cable Transition for 3D Electromagnetic Simulation by Equivalent Circuit Assisted De-Embedding

Herbert Hackl, Bernhard Auinger, Mate Kovacs, Andreas Wagner, Christian Stockreiter

Research output: Conference proceeding/Chapter in Book/Report/Conference Paperpeer-review

Abstract

This paper describes a procedure to generate a 3D model for electromagnetic simulation of the transition region, i.e. solder connection, between a printed circuit board (PCB) trace and a coaxial cable. The PCB-to-coax launch cannot be directly accessed by measurement, thus de-embedding is used to extract the section under investigation from measurements. However, de-embedding procedures like 2xThru, that require fixtures with low variances, are not suitable for this task due to considerable variation in solder joints and available test structures. Instead, de-embedding based on an equivalent circuit is proposed. The circuit is partitioned into physical meaningful sections, such that the 3D model of the targeted PCB-to-coax interface can be verified by comparison to respective parts of the circuit. S-parameters up to 26.5 GHz and time domain reflectometry (TDR) are used to validate the simulation model against measurements of multiple hardware samples.
Original languageEnglish
Title of host publication2022 International Symposium on Electromagnetic Compatibility – EMC Europe
Pages372-376
Number of pages5
DOIs
Publication statusPublished - 8 Sept 2022
Event2022 International Symposium on Electromagnetic Compatibility – EMC Europe - Gothenburg, Sweden
Duration: 5 Sept 20228 Sept 2022

Conference

Conference2022 International Symposium on Electromagnetic Compatibility – EMC Europe
Period5/09/228/09/22

Keywords

  • Solid modeling
  • Three-dimensional displays
  • Coaxial cables
  • Fixtures
  • Reflectometry
  • Electromagnetic compatibility
  • Frequency measurement

Fingerprint

Dive into the research topics of 'Measurement-Based Modeling of PCB-to-Coaxial Cable Transition for 3D Electromagnetic Simulation by Equivalent Circuit Assisted De-Embedding'. Together they form a unique fingerprint.

Cite this