Microwave Characterization of Low-loss FDM 3-D Printed ABS with Dielectric-filled Metal-pipe Rectangular Waveguide Spectroscopy

2019 
Over time, the accuracy and speed by which a material can be characterized should improve. Today, the Nicolson–Ross–Weir (NRW) methodology represents a well-established method for extracting complex dielectric properties at microwave frequencies, with the use of a modern vector network analyzer. However, as will be seen, this approach suffers from three fundamental limitations to accuracy. Challenging NRW methods requires a methodical and robust investigation. To this end, using a dielectric-filled metal-pipe rectangular waveguide, five independent approaches are employed to accurately characterize the sample at the Fabry–Perot resonance frequency (non-frequency dispersive modeling). In addition, manual Graphical and automated Renormalization spectroscopic approaches are introduced for the first time in the waveguide. The results from these various modeling strategies are then compared and contrasted to NRW approaches. As a timely exemplar, 3-D printed acrylonitrile–butadiene–styrene (ABS) samples are characterized and the results are compared with existing data available in the open literature. It is found that the various Fabry–Perot resonance model results all agree with one another and validate the two new spectroscopic approaches; in doing so, exposing three limitations of the NRW methods. It is also shown that extracted dielectric properties for ABS differ from previously reported results and reasons for this are discussed. From measurement noise resilience analysis, a methodology is presented for determining the upper bound signal-to-noise ratio for the vector network analyzer (not normally associated with such instrumentation). Finally, fused deposition modeling (FDM) 3-D printing can result in a non-homogeneous sample that excites open-box mode resonances. This phenomenon is investigated for the first time analytically and with various modeling strategies.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    17
    References
    7
    Citations
    NaN
    KQI
    []