Nondestructive and noncontact dielectric measurement methods for low-loss and high-loss liquids using free space microwave measurement system in 8-12.5GHz Frequency range / Mohd Aziz Aris
Knowledge of wideband dielectric properties of liquid materials is necessary in many applications such as biomedical, remote sensing, powder technology and radar absorbing materials. Nondestructive, noncontact, in situ and real time measurement of dielectric properties of liquids is important for...
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English |
| Published: |
2005
|
| Subjects: | |
| Online Access: | http://ir.uitm.edu.my/id/eprint/3604/ http://ir.uitm.edu.my/id/eprint/3604/1/TM_MOHD%20AZIZ%20ARIS%20EE%2005_5%201.pdf |
| Summary: | Knowledge of wideband dielectric properties of liquid materials is necessary
in many applications such as biomedical, remote sensing, powder technology and
radar absorbing materials. Nondestructive, noncontact, in situ and real time
measurement of dielectric properties of liquids is important for evaluation of
complex material systems such as service-aged transformer oil. Free-space
microwave measurement (FSMM) system (which is nondestructive and noncontact)
was developed for accurate measurement of dielectric properties of low-loss and
high-loss liquids at microwave frequencies. Three free-space methods were used,
namely, reflection and transmission method, transmission only method and metalback
method. These methods were especially developed and tested for dielectric
measurement of liquids in free-space. This is the first reported implementation of
transmissions only and metal-back methods in free-space for low-loss liquids at
microwave frequencies. FSMM system consists of spot focusing of hom lens
antennas, mode transitions, coaxial cables and vector network analyzer (VNA). All
measurements were made using FSMM system in the frequency range of 8 GHz to
12 GHz and temperature fixed at20° C . Inaccuracies due to diffraction from the
sample are minimized by using spot focusing hom lens antennas. Errors due to
multiple reflections between antennas were minimized by using free-space LRL
(line, reflect, line) calibration technique and time domain gating (a feature ofVNA).
The liquid is contained in a container consisting of two Plexiglas plates with known
dielectric properties. These Plexiglas plates are designed to be quarter wavelength at
mid band to reduce reflectivity of the liquid sample. The effect of Plexiglas plates is
removed from the knowledge of dielectric properties of Plexiglas and thickness of
plates. Dielectric properties of low-loss liquids such as Dioxane, Benzene and
transformer oil are measured using reflection and transmission, transmission only
and metal-back methods. High-loss liquids such as Methanol, Ethanol, Ethylene
Glycol and N-Propyl Alcohol, dielectric constants and loss factors were measured
using reflection and transmission, transmission only and metal-back methods. It was
found that reflection and transmission method gives accurate results for high-loss
liquids. Transmission only and metal-back methods give accurate results for lowloss
liquids. Values of dielectric constant for low-loss liquids are almost constant in '
the frequency variation. Whereas, dielectric constant of high-loss liquids are reduce
in magnitude if frequency increased. Results for all liquids are compared with
published data. For high-loss liquids, our experimental results are also compared
with data calculated from previo.usly published Debye parameters. All results for
dielectric constants and loss factors for all samples have close match with published
data. The percentage error for dieleetric constant is ±3.9% and for loss tangent is
± 0.008 in magnitude for metal-back method. The transmission only method gives
an error of 3.83% for dielectric constant and ± 0.02 for loss tangent. The percentage
error for dielectric constant is ± 2.9% and error for loss tangent is ± 0.025 in,
magnitude for reflection and transmission method. These errors in dielectric
properties are due to error in measurement of S- parameters using FSMM system.
These errors in S-parameter measurements are due to post-calibration systematical,
random and drift errors. |
|---|