Note: Descriptions are shown in the official language in which they were submitted.
127i~2A
72497-1
This invention relates to apparatus for measuring the
mean permittivity (dielectric constant) of a sample of material.
A measurement oE the mean permittivity of a sample of
material in a chamber enables one to estimate the volume ratio of
two materials of known properties making up the sample,at least
one of the two materials being a fluid, i.e. gas or liquid. One
can also measure the dielectric constant or permittivity of an
unknown material of known volume when surrounded by a fluid of
known properties. In all cases, the materials under test may be
either static or in motion as for a contained flow situation.
Canadian patent No. 1,082,310 of Dechene, et al, issued
July 22, 1980 discloses apparatus for measuriny relative fractions
of liquid vapor in mixed phase fluid flow. Measurements are made
in several current loops and the current loops are supplied by a
rotating electrical field source by means of multi-phase voltage
oscillations of 1-30 kHz. A reference sensor is used and the
voltage across a load resistor represents conductivity.
Canadian patent No. 1,101,070 of Newton, et al, issued
May 12, 1981, relates to measuring relative fractions of liquid
(nonconductive) and vapor, or solids (nonconductive) and gases.
Sequential capacitance measurements are made across the cross
section of a flow to be measured. Frequencies of 10-100 kHz are
used and a rota-ting electric field is produced to average the
dielectric constant of the entire cross section of the sensor.
U.S. patent 3,639,835 of Dammig, Jr., et al, issued
February 1, 1972, discloses a capacitive tank gaging apparatus
~5~
72497-1
based on the fact that the density of a dielectric liquid is
directly related to its dielectric constant. The capacitance
between electrodes separated by space containing liquid and gas of
known characteristics is a measure of the quantity of liquid
present. A measuring unit impresses different voltages on sets of
electrodes in a manner that generates relatively uniform time-
varying electric fields across the tank. The measuring unit
senses the capacitive current -through the tank interior between
the electrodes, which current is a function of the dielectric
corlstan-ts of -the fluids in the -tank.
The present invention, which differs in a number of
respects from the arrangements disclosed in -the above mentioned
patents, enables simple and highly accurate estimation of the mean
permittivity of the contents of a sense chamber. Briefly, the
invention uses a sense chamber consisting of a cylindrical tube of
relatively low permittivity, low conductivity material (e.g. poly-
tetrafluoroethylene). This is wrapped with a plurality of helical
foil sense electrodes, e.g. three, each covering about 30% of the
tube circumference. A thin insulating barrier surrounds the sense
electrodes. On top of the insulating barrier a set of foil shield
electrodes is wrapped to overlap the sense electrodes, these
extending slightly beyond the edges of the sense electrodes. One
more insulating barrier surrounds the shield electrodes and a
continuous grounded Faraday shield surrounds the entire sense
chamber except for an opening at the top through which sample
material may be poured.
72497-1
Three sinusoidal signals of the same frequency, e.g. 10
kHz, and 120 of phase apart are applied to the sense and shield
electrodes such that approximately the same voltage is applied to
each side of a pair of electrodes. This results in a rotating
electric field. A reading signal is developed which is a voltage
proportional to the sum of the true RMS value of the charging
currents applied to each of the three sense electrodes.
The ratio of the signal output span (difference between
full and empty readings) with a known subs-tance and with a test
sample i5 nearly proportional to the ratio of the permittivities
of the two substances.
The arrangement according to the above mentioned
Canadian patent No. 1,082,310 is similar in use of a three-phase
rotating field but differs in that the fluid is in contact with
the electrodes so that current can flow through the fluid. In the
present invention there is no electrical contact with the sensed
materials and measurements are not reliant on conductivity
effects. In addition, the present invention requires only three
electrodes whereas the system according to the patent requires six
electrodes~ The system of the patent uses alternate grounded
electrodes providing a more complex field pattern and it lacks the
driven guard electrodes of the present invention which ensure that
the electric field set up by the sense electrodes is virtually
exclusively within the volume of the chamber. Should the system
of the patent be used in a capacitance mode, a great deal of
sensitivity would be lost due to fringed field effects and current
~i2~L
72497-1
required to polarize the ma-terials behind and beside the
electrodes.
Regarding Canadian patent No. 1,101,070, although it
relates to a capacitance system it uses a distinctly different
field geometry using six electrodes as opposed to three for the
preferred embodiment of the present invention, and a complicated
switching system is required to obtain rotation of the field. As
with Canadian patent No. 1,082,310, the electrode plates are in
intimate con-tact with the sample fluid and therefore the system
would be quite sensitive to conductivity effects, i.e. readings
would change based on ionic content of water, for instance. The
system of the presen-t invention is nearly insensitive to dissolved
salts in the water. The superior guard and shield arrangement of
the present invention improves the sensitivity as there need be no
energy input by sense electrodes to polarize materials outside or
adjacent to these electrodes.
Regarding U.S. patent ~o. 3,639,835, the concept of the
capacitance gauging system is correct but it does not use a
rotating field and, due to the distinctly patchwork nature of
the electrodes within the chamber, it is unlikely that the fields
are uniform, nor would the time average of the fields be likely to
be uniform throughout the volume of the chamber. Therefore,
inaccuracies may occur for situations in which fluids within the
tank had sloshed to one side or the other or towards the ends.
The patent requires a considerably more complicated network than
the present invention in order to obtain readings. With respect
to the flow through system of the patent, -there is no helical
7~497-1
electrode arrangement which would reduce sensitivity to non-homo-
geneous fluid flows, nor is the selected electrode arrangement
adequate to produce a rotating electrical field. In all cases,
the patent is concerned with a static single phase AC field.
According to a broad aspect of the present invention
there is provided apparatus for measuring the mean permittivity of
a sample of material comprising a sense chamber for containing
said sample, said sense chamber comprising a cylindrical tube of
low permittivity, low conductivity material and having an outer
surface provided with at least three helical foil electrode
structures extending along substantially the entire length of said
tube, each electrode structure comprising an inner sense electrode
in contact with said tube and an outer shield electrode, said
sense electrode and said shield electrode being separated by a
thin barrier of insulating material, said apparatus further
comprising means for applying out of phase alternating voltage
signals to said electrode structures to produce a rotating
electrical field in said sense chamber and for deriving a voltage
read-out signal proportional to -the sum of the charging currents
applied to the sense electrodes of said electrode structures, said
read-out signal being directly related to the mean permittivity of
said sample.
An embodiment of the invention will now be described in
con~unction with the accompanying drawings, in which:
Figure 1 is a sketch of a sense chamber provided with
three single-turn helical electrode structures,
72497-1
Figure 2 is a cross-sectional view of a sense chamber
and associated circuitry with radial spacings exaggerated for
clarity of illustration, and
Figure 3 is a block diagram of drive and sense circuitry
used in the apparatus according to the invention.
Referring to Figure 1, the sense chamber 10 comprises a
cylindrical tube 12 of relatively low permittivity, low conduc-
tivity material, e.g. polytetrafluoroethylene (sold under the
trade mark TEFLON). The tube 12 is wrapped with three helical
electrode structures 13, 14 and 15 which, as shown in Figure 2,
each comprises a foil sense electrode 16, a thin insulating
barrier 18 and a foil shield electrode 20. Each sense electrode
16 has a width sufficient to cover about 30~ of the tube circum-
ference. The remaining 10~ of the circumference is taken up by
gaps between electrodes, one of which is indica-ted at 21 in
Figure 2. The insulating barrier 18 surrounds the sense
electrodes and covers the gaps between them. The shield
electrodes 20 overlap the sense electrodes 16 and are wider so as
to extend beyond the edges of the sense electrodes by about 25~ of
the gap width on each side. Another insulating barrier 24
surrounds the shield electrodes and a continuous grounded Faraday
shield 26 surrounds the entire sense chamber except for an opening
at the top through which sample may be poured.
Drive and sense circuitry 30, to be described in connec-
tion with Figure 3, is connected to the shield and sense Eoils of
each electrode structure via lines 32 and 3~. It is important to
note that, while Figure 2 shows separate sets of lines 32, 34, for
72497-1
ease of illus-tration, the lines to the shie].d and sense electrodes
are actually coaxial. Thus each sense electrode will be connected
to circuitry 30 by the center conductor o~ a coaxial cable while
the associated shield electrode will be connected by the shield
conductor of the same coaxial cableO This ensures that the only
capacitive effects are at the sense electrodes. Parallel lines or
twisted pairs would have stray capacitances which might well
exceed the small capacitance effects at the sense electrodes so
that the apparatus would not function properly.
Turning now to Figures 3, a master oscillator 40
produces a sinusoidal signal of, for example, 10 kHz, which is fed
to three phase shifters 42, 43 and 44. The phase shifters are
adjusted to produce three signals of identical frequency but 120
of phase apart. The outputs of the phase shifters are connected
to shield d~iving amplifiers 47, 48 and 49 having output lines 32a
to 32c connected to the shield electrodes 20 (not shown in
Figure 3). The outputs of the phase shi~ters 42 to 44 are also
connected via matched resistors 50a to 50c and lines 51a to 51c
(actually coaxial shields for lines 32a to 32c) to the sense
electrodes 16 (Figure 2.) ~early the same voltage is applied to
the sense and shield electrodes resulting in a rotating electric
field. Charging curren-ts to the sense electrodes, dependent on
the permittivity of the material in the sample chamber, develop
reading voltages or sense voltages across resis-tors 50a to 50c
which are amplified by amplifiers 54 to 56 and conver-ted to DC
signals by signal converters (rectifiers) 60-62. The DC signals
æ~
72497-1
from converters 60-62 are summed by summation stage 65 to produce
an output Vout which is a voltage proportional to the -true
RMS value oE the charging currents applied to each of the three
sense electrodes.
The use of helical electrodes results in good linearity
of measurements and reduces the sensitivity to orientation for
nonhomogeneous flowing systems.
The sample chamber could be a section of tubing with
fluid material flowing through it rather than a "static" sample
chamber as described above.
The driven cable shields and guard electrodes maximize
sensitivity by minimizing current in the sense lines and
electrodes required to polarize materials other than those within
the sense chamber.
Although Figure 1 shows single turn helical electrodes
they could be, for example, double turn helices. More than three
sets of electrodes can be used and the driving signals need not be
sinusoidal.
-- 8 --
