Frequently, when free-space electromagnetic waves pass through a material, there will be some form of interaction between the wave and the material. Measuring this change forms the basis of free-space, dielectric material measurement, where the variations will be attenuation and a phase change relative to the wave when the material was not present and are typically recorded over a broadband range of frequencies. In this work a new technique is presented to accurately perform free-space broadband material measurement to calculate the dielectric response of materials of unknown electrical length. (In this thesis, the electrical length means that while the physical length of the path is known, the number of wavelengths in the material is unknown, due to the unknown permittivity. This unknown number of wavelengths is described as the electrical length both here and also in many other works.) The technique is eventually applied to a more difficult form of material to measure, which is material in the form of powders and liquids. Powders and liquids present a particularly difficult challenge in that they must also be contained in a container of uniform shape and sufficient cross-section to guarantee that the entire received wave has passed through the same amount of material. The technique presented works for containment vessels with no or some dielectric response (respectively being relatively easy analysis and very difficult for dielectric analysis). Much of the work is devoted to the development of a test infrastructure and also a new vector network analyser (VNA) calibration technique that would be more appropriate to this infrastructure and kind of measurement than existing techniques. In terms of the latter point, to minimise measurement errors a free-space, broadband Thru-Reflect-Line/Match (TRL/M) calibration technique was developed. The challenge of a test structure for materials with an unknown electric length was met by offering the ability to vary the thickness of material through which the probing wave passes. Moreover, the range over which the thickness could vary was quite large (3 mm up to 18 cm) to accommodate materials of unknown electrical length with dielectric constant values over a wide range (2 to 80). Two measurement techniques are presented based on horizontal and vertical probing wave propagation, both of which allow for a series of measurements to be taken where the varying parameter is thickness of material under test. However, the horizontal propagation measurements required an extra containment layer (to contain the powder/liquid on both sides normal to the propagation, rather like a large, slim fish tank), compared to the vertical propagation measurement, but had significant mechanical and also the computational difficulties mentioned earlier. The vertical propagation measurements on the other hand show more promise, both in terms of mechanical containment and also in the inversion of the measured S-parameters to arrive at the complex electric permittivity. The measurements were performed in a large hall and also in a Radio Frequency (RF) anechoic chamber (optimal test range of approximately 500 MHz – 18 GHz). Several diverse solid materials were tested initially in developing the test infrastructure including double glazed standard and low emissivity windows, common building materials, house insulation materials and automobile glass. Once the platform and calibration technique were competed to a satisfactory level, measurements intended for dielectric analysis were carried out on commercial glass panes, water and household plain flour. All measurements were subsequently published in various conferences and workshops, hosted by the IEEE, IEEE Communications Society, IET, the European Association on Antennas and Propagation (EurAAP) and the Royal Irish Academy’s research Colloquium on Communications and Radio Sciences workshop.
|Publication status||Unpublished - 2014|
- Broadband, Dielectric measurement techniques