Note: Descriptions are shown in the official language in which they were submitted.
CA 02334063 2001-02-02
1
SYSTEM AND METHOD FOR MEASURING
LIQUID METAL LEVELS OR THE LIKE
BACKGROUND
In a casting process, the liquid metal needs at one point to be transferred
from an
holding furnace to the molds of a casting pit, where it is poured into the
molds and
cooled to make ingots or billets. The transfer of liquid metal from the
holding
furnace to the molds is generally made using an opened or closed channel
called
a launder. A launder is also called a supply gutter in some other references.
A
launder may further be used for transferring liquid metal from an alloying
furnace,
if any, to the holding furnace. As its name indicates, an alloying furnace is
used
for combining various metals together in the required proportions so as to
prepare
alloys.
The exterior walls of a launder are usually made of mild steel and constitute
the
frame thereof. The interior side of the frame is generally lined with a layer
of
compacted ceramic wool or another kind of resilient and high-temperature
resistant insulating material. The portion of the launder in contact with the
liquid
metal is typically made of a solid refractory material. The refractory
material is
used to reduce the heat losses and to prevent the pick-up of contaminating
materials.
The holding furnace typically contains several tons of liquid metal which need
to
be transferred to the casting pits over a period of time ranging from a few
hours in
the case of a semi-continuous process, to many consecutive days in the case of
a
continuous process. A key factor for the full success of a casting operation
is the
uninterrupted and constant supply of liquid metal during the transfer. If the
metal
stops from flowing or if the flow rate changes while the casting operation is
under
way, appropriate actions and corrective measures have to be taken immediately.
As a result, the transfer and casting operation require that the level of
liquid metal
flowing through the launder be measured and monitored in a reliable and
accurate
CA 02334063 2001-02-02
2
fashion. There is thus a need for a system to continuously monitor the level
of
liquid metal in a launder so as to ensure that the proper amount is
continuously
flowing.
While some prior attempts to provide devices for measuring the level of liquid
metal in a launder have resulted in a number of different constructions, none
has
been found completely satisfactory. For instance, some systems use a laser
beam to measure the reflectivity of the launder and its contents. These
systems
use the surface reflectivity to obtain a signal back from the liquid metal and
to
measure the level thereof. However, when the surface is too shiny or when the
dross is too thick, the signal is lost. Similar problems may happen when dense
fumes obscure the region above the launder. Some other systems use the
electrical capacitance of the launder and its contents. However, there is a
suitable
response only for a few inches and this is generally not sufficiently precise
nor
accurate in many applications. There is also a low signal to noise ratio,
making it
difficult to obtain an accurate value of the liquid metal level.
SUMMARY
The object of the present invention is to reduce the difficulties and
disadvantages
experienced with prior systems by providing an improved system and a method
for
measuring liquid metal levels in a launder or any similar locations where such
measurements need to be undertaken. An important aspect of the present
invention is that it is not significantly affected by the presence of steel on
the
exterior side of the launder. It is further stable in the harsh environment of
a cast
house and may work even if there is no external cooling.
The full extent of the present invention will be more readily apparent from
the
following detailed description of preferred embodiments thereof, which
proceeds
with reference to the accompanying figures.
CA 02334063 2001-02-02
3
BRIEF DESCRIPTION OF THE FIGURES
FIG. 1 is a perspective view of a typical launder with a probe mounted on the
side
thereof, in accordance with a first and preferred embodiment of the present
invention.
FIG. 2 is an exploded perspective view of the probe with the launder
illustrated in
FIG. 1, showing the aperture over which the probe is mounted.
FIG. 3 is a perspective view of a typical launder with a probe in an overlying
disposition in accordance with another possible embodiment of the present
invention.
FIG. 4 is a block diagram showing the preferred connections between the
components.
FIG. 5 is a cross-sectional view of the probe and the frame of the launder, as
taken along line 5-5 in FIG. 1.
FIG. 6 is an exploded perspective view of some of the parts of the sensor
portion
of the probe, in accordance with a preferred embodiment of the present
invention.
FIG. 7 is a view similar to FIG. 6, showing the parts from another angle.
FIG. 8 is a cross-sectional view of a double-sheathed wire, in accordance with
another aspect of the present invention.
FIG. 9 is a graph showing the effect of the temperature inside the receptacle
of the
probe on the signal sensed by the receiver coil.
CA 02334063 2001-02-02
4
FIG. 10 is a graph showing an example of the relationship between the signal
in
the receiver coil and the level of metal in the launder.
FIG. 11 shows two superposed graphs representing an example of the level of
liquid metal as measured with a device found in the prior art (A) and with a
system according to the present invention (B).
DETAILED DESCRIPTION
A system (10) according to a possible and preferred embodiment of the present
invention is described hereinafter and illustrated in the accompanying
figures.
Throughout these figures, analogous components are identified by the same
reference numerals.
FIGS. 1 to 3 show an example of a typical launder (12) which constitute the
main
location where the present invention can be used. However, it should be
understood that the present invention is not limited for its use with a
launder and
can thus be used elsewhere. For instance, it can be used in conjunction with
molds, electrolytic cells or any other suitable location.
As aforesaid, the exterior walls (14) of a launder (12) are usually made from
mild
steel and constitute the frame thereof. The interior of the launder (12) is
lined with
a layer of compacted ceramic wool (16). The portion of the launder (12) in
contact
with the liquid metal is made of a solid refractory material (18). When a
liquid
metal flows in the launder (12), the temperature of the frame (14) typically
increases from room temperature to about 200°C in the case of aluminum.
Since
the refractory material (18) and the frame (14) do not have the same thermal
expansion coefficient, they expand at different rates, creating a relative
displacement between them. The intermediate layer of compacted ceramic wool
(16) allows the dissimilar expansions to be compensated, thereby ensuring that
the refractory material (18) be held properly in position throughout all the
range of
CA 02334063 2001-02-02
temperatures. It also provides some thermal insulation in addition to that
provided
by the refractory material (18).
The system (10) comprises a probe (20) which is held close to the location
where
5 the liquid metal is present. The probe (20) essentially comprises a sensor
assembly and a fastening assembly. FIG. 1 shows how the exterior parts of the
fastening assembly are held against one of the walls (14) of the launder (12).
It
should be noted that it is also possible to mount the probe (20) above the
launder
(12) instead of mounting it on a lateral side thereof. FIG. 2 shows that the
probe
(20) in FIG. 1 is held over an aperture (22) made through the frame (14) of
the
launder (12). This aperture (22) is either cut using a torch or a saw for
instance, or
is a part of a launder designed for that purpose. The removed section is as
small
as possible so as to prevent the structure from weakening. A typical width for
the
aperture (22) is 168 mm. The ceramic wool (16) is also removed from the
aperture (22) so as to expose the outer side of the refractory material (18).
FIG. 3 shows another possible embodiment of the present invention. This
embodiment is characterized in that the probe (20) is located over the center
of the
launder (12). The probe (20) is held in place using appropriate fasteners, as
apparent to a person skilled in the art. Example of fasteners for this purpose
include brackets, rods, plates and/or others, all of which are located as far
as
possible from the probe (20) or made of a material having no significant
effect on a
magnetic flux.
The present invention is based on electrical inductance. Inductance is the
phenomenon where a changing electrical current in one electrical circuit
builds a
magnetic field which is capable of inducing an electromagnetic force and an
opposing current in an adjacent circuit. These circuits are in the form of
coils in
the present invention.
CA 02334063 2001-02-02
6
FIG. 4 schematically represents the preferred connections of the electrical
components. It illustrates the two coils (30,32) having a side-by-side
configuration,
which means that they are on a same side but are spaced-apart from each other.
One coil is an emitter coil (30) and the other is a receiver coil (32). Each
coil
(30,32) comprises a wire wound around itself numerous times, preferably around
the edge of a respective core element (34,36). The wire winding is made along
the length of the core elements (34,36). These core elements (34,36) are
preferably in the form of plates, but other forms or shapes are also possible.
They
could also be slightly concave or convex instead of being flat plates.
The core elements (34,36) are made of a material having a high magnetic
permeability and which can transmit the magnetic flux towards the interior of
the
launder (12) where the measurements are taken. Such material should have a B/H
value, representing the magnetic permeability, between 10 000 and 1 000 000.
It
also has to resist the temperatures reached inside the probe (20). An example
of
a suitable material is the one known as MuMetaIT"~ or Hy Mu 80T"", and which
typically contains nickel (80%), iron (15%), Molybdenum (4,2%), Manganese
(0,5%) and carbon (0,02%). Heat-treating the alloy in dry hydrogen to increase
the grain size enhances the magnetic properties of the material. Other
materials
can be used as well, such as the ones known as Magnifier 7904T"",
PermaIloyT"",
HypernomT"', etc.
Preferably, the core elements (34,36) are disposed such as to have their
longitudinal axis being somewhat perpendicular to the flow direction of liquid
metal. However, each of the core elements (34,36) can be set at an angle which
varies from about 30 degrees on both sides of a perpendicular position.
Briefly stated, the emitter coil (30) is used to generate an alternating
magnetic flux.
Both coils (30,32) are arranged and disposed in a way that the alternating
magnetic flux from the emitter coil (30) induces a voltage signal in the
receiver coil
(32). They are also held in close proximity of the location of liquid metal.
In the
CA 02334063 2001-02-02
7
case of the launder (12), the coils (30,32) are held close against the side of
the
refractory material (18). The magnetic flux is carried through the refractory
material (18) and then across the path of liquid metal. There is thus no
direct
contact between the coils (30,32) and the liquid metal, the system (10)
working
completely in a remote manner.
The receiver coil (32) continuously receives a signal from the emitter coil
(30) even
if the launder (12) is empty. However, the overall signal through the receiver
coil
(32) increases in presence of liquid metal. This changes the signal measured
in
the receiver coil (32) and ultimately allows the system (10) to determine the
level
of liquid metal upon analysis of the variation of the signal measured in the
receiver
coil (32)
The present invention can be used with a very wide range of metals, including
and
not limited to aluminum, brass, copper, iron, lead, magnesium, steel,
titanium, zinc
and many others, or their alloys. Furthermore, although the system (10) is
primarily intended for use with liquid metals, it can also be used with melted
salts
that are electric conductors.
Referring to FIG. 5, there is shown a cross-sectional view taken from the top
of the
launder (12) illustrated in FIG.1. It shows the sensor assembly, which is
located
inside the probe (20). In addition to the coils (30,32), the sensor assembly
comprises a two-part hollow receptacle (40), which is designed to hold and
protect
the two coils (30,32). Both parts (40a,40b) of the receptacle (40) are made of
a
ceramic material or any other suitable material having no or only a weak
effect on
an electromagnetic signal to be sent from the emitter coil (30) to the
receiver coil
(32).
FIGS. 6 and 7 show the two parts (40a,40b) of the receptacle (40) being
separated
from each other. They are preferably made of a carbon-carbon composite, such
as the one known under the trade name "K-Karb" from Kaiser Compositek. It has
CA 02334063 2001-02-02
been found that this material has the required mechanical properties at high
temperature and no adverse effect on the signal. It is further capable of
withstanding the harsh environment of a cast house. Other materials, such as
alumina, silica, mullite, any combinaison of alumina with silica or zirconia,
are also
suitable candidates for the construction of the receptacle (40). The
receptacle (40)
is provided with a hollow internal housing (40c). Slots (41 ) are provided in
the
second part (40b) to hold the corresponding core elements (34,36). The slot in
the
middle is used for an optional third core element (60), which is described
later.
Advantageously, the middle slot is provided with a hole (41 a) to accommodate
the
wires that need to reach the housing (40c).
As best shown in FIG. 5, there is also provided a temperature sensor (42)
within
the housing (40c) in order to measure the temperature in the vicinity of the
coils
(30,32). The temperature sensor (42) preferably comprises either a
thermocouple
or a resistivity thermal device (RTD). The selection between the two kinds of
temperature sensors is essentially dependent upon the highest temperature
reached in the housing (40c). A RTD is preferred when the temperature is about
400°C or less since it is less expensive than a thermocouple. It should
be noted
that the electrical wires have been omitted from FIG. 5 for clarity purposes.
Each
coil (30,32) has two electrical wires and the temperature sensor (42) has also
two.
A total of six electrical wires are coming out of the sensor assembly.
The wire used in constructing the coils (30,32) has to be electrically
insulated but
the interior portion has to be a good electrical conductor. It further has to
resist to
high temperature oxidation. Hence, copper can not be used alone since it
rapidly
losses its electrical conductivity as it becomes oxidized due to in the
environment
and high temperatures in cast houses.
When the temperature inside the probe (20) does not exceed about
400°C, a
nickel-clad copper or aluminum wire can be used. The nickel-clad copper wire
preferably has a diameter between 0,15 and 1,0 mm. A wire made of aluminum
CA 02334063 2001-02-02
9
should have a purity of 99,5% or higher in order to be a good conductor. The
aluminum wire preferably has a diameter between 0,25 and 1,5 mm in diameter,
the most preferred diameter being between 0,5 and 0,8 mm. The electrical
insulator covering the wire may be a glass or mica sheath. For aluminum,
alumina
obtained by anodization could be used. A very suitable form of anodized
aluminum is the one commercially obtained from Alumat Inc. (Ponoma,
California),
which allows the wire to be shaped without breaking the layer of alumina.
Other
materials can be used as well.
When the temperature inside the probe could exceed about 400°C, the
wires used
in the coils (30,32) have to be designed to withstand high temperatures. This
could be achieved using a double-sheathed wire (50) made in accordance with
another aspect of the present invention. FIG. 8 shows an example of a cross
section of this double-sheathed wire (50). It could be prepared in one step by
a
conventional cold drawing system.
The double-sheathed wire (50) preferably comprises a copper or pure nickel
core
(52). The core (52) comprises a first seamless sheath (54) made of a high
nickel-
base content alloy or other malleable non-oxidizing alloys. Preferably, the
first
sheath (54) is made of an InconeITM 600 alloy. The core (52) and the first
sheath
(54) are enclosed within a second sheath (56), preferably made of the same
material as the first sheath (54). These components are coaxially disposed and
are further provided with an annular space (58) between them. This space (58)
is
preferably packed with an electrically insulting and high temperature
resistant
material to prevent them from being in electrical contact. This material is
preferably a ceramic powder, such as magnesium oxide (Mg0). Other materials
could be used as well.
The double-sheathed wire (50) preferably has an outside diameter between 0,8
and 2,0 mm, most preferably between 1,0 and 1,5 mm. The inside diameter of the
second sheath (56) is preferably between 0,4 and 1,6 mm, most preferably
CA 02334063 2001-02-02
between 0,6 and 1,1 mm. Typically, the resistivity of a 1,0 mm outside
diameter
double-sheathed wire (50) with a nickel clad-copper core is about 1,7 milliohm
per
cm at room temperature. This increases to about 10 milliohms per cm at
1100°C.
It has been found that this wire (50) can be used for several weeks in
oxidizing
5 atmospheres having a temperature up to 1100°C. It should be noted
that the
double-sheathed wire (50) could also be used in other high temperature
applications.
Generally, the length of the core elements (34,36) determines the height of
liquid
10 metal that can be measured. For example, core elements (34,36) having 150
mm
in length can measure between 1 and 150 mm of liquid metal. Core elements
(34,36) typically can be made from 5 mm to 800 mm in length and have windings
between 30 and 200 turns. The preferred number of turns is between 60 and 120,
the most preferred being about 90.
The thickness of the core elements (34,36) is also an important parameter.
Thick
coils produce a magnetic field that occupies a large volume. If the magnetic
field
is too wide, then it would go through the steel frame (14) of the launder
(12). This
is undesirable since the temperature of the frame (14) changes drastically
during a
cast, causing a signal drift. For example, during a normal cast of aluminum,
the
temperature of the frame (14) of the launder (12) increases from room
temperature
to 200°C.
While reducing the thickness of the core elements (34,36) decreases the part
of
the magnetic field going through the frame (14), it has the drawback of
decreasing
the signal strength. This undesirable effect is due to the fact that the
magnetic
field created on one side of the emitter coil (30) is opposite the one created
by the
other side thereof. The result is that the magnetic field transmitted to the
receiver
coil (32) would have a lower intensity. The first (34) and the second (36)
core
elements are allowing to solve this problem. Moreover, for keeping an even
higher
intense and concentrated field, a third core element (60) may be placed
between
CA 02334063 2001-02-02
11
the two coils (30,32). The third core element (60) is preferably made of the
same
material than that of the other core elements (34,36) or an equivalent. It
contributes to further focussing the magnetic field through the center
aperture (22),
thereby reducing its interaction with the frame (14) of the launder (12).
Spacers
(49) are preferably used to maintain the spacing between all core elements
(34,36,60).
Preferably, as shown in FIGS. 5 to 7, the third core element (60) is placed
perpendicular to the coils axis. It is in the form of a sheet having between
about
50 and 280 mm in length, between about 15 and 50 mm in height, and between
about 0,1 and 5 mm in thickness. With the third core element (60), the
distance
between the two coils (30,32) is typically from 5 to 30 mm, depending on their
length, compared to between 50 and 75 mm without a third core element. The
third core element (60) also enables the system (10) to measure the effect of
the
inductance in the liquid metal at a distance of up to 100 mm away from the
probe
(20) into the launder (12). Concentrating the magnetic flux at the center of
the
aperture (22) decreases the distance between the coils (30,32) and reduces the
portion of magnetic flux going through the frame (14) of the launder (12). The
effect of the surrounding steel thus becomes negligible when the third core
element (60) is placed between the two coils (30,32).
As aforesaid, the probe (20) needs to be held in place while it is used. It is
necessary that the sensor assembly of the probe (20) be held so that the
distance
between the coils (30,32) and the liquid metal does not change. The contrary
would cause a signal drift and thus give an incorrect indication of the liquid
level.
When the probe (20) is above the launder (12), there is no significant change
throughout the use of the system (10). However, when the probe (20) is
installed
on the frame (14) of the launder (12), the fastening of the probe (20)
requires
some attention because the relative distance between the frame (14) and the
center of the launder (12) changes with the thermal expansion.
CA 02334063 2001-02-02
12
To solve the above-mentioned problem, a novel fastening assembly (90) has been
devised so as to allow the sensor portion of the probe (20), and thereby the
coils
(30,32), to be held at a constant distance from the outer side of the
refractory
material (18). This keeps the distance between the coils (30,32) and the
liquid
metal as constant as possible over the range of temperatures.
The fastening assembly (90) preferably comprises a protective cover (92) that
is
made of a ceramic material or a carbon-carbon composite material. Other
suitable
materials can be used as well. The cover (92) is removably mounted around the
aperture (22) made through the frame (14) of the launder (12). To achieve
this,
the fastening assembly (90) preferably comprises a fixation frame (94) welded
around the aperture (22). This fixation frame (94), shown in FIG. 2, is
preferably
made of stainless steel 300 series and comprises fastening bolts (94a)
projecting
therefrom. Other materials can be used as well. Referring now to FIG. 5, the
protective cover (92) is inserted over the fixation frame (94) and the bolts
(94a) are
inserted through corresponding holes in the cover (92). The free end of the
bolts
(94a) protrudes from the exterior of the cover (92) and nuts (94b) are used to
lock
the cover (92) in position.
The central portion of the cover (92) is preferably provided with three holes.
One
is to accommodate a tube (96) through which the electrical wires will run.
FIGS. 1
to 3 and 5 show that the tube (96) ends with an enlarged adapter (98) in which
the
terminals of the electrical wires of the probe (20) are connected to
corresponding
external wires. The other side holes (101 ) are receiving corresponding bolts
(100).
The tube (96) and the bolts (100) are free to slide in their respective hole.
These
bolts (100) are preferably made of stainless steel but made be made of other
suitable alloys. One end of the bolts (100) is threaded and is located in
corresponding chamfers (102) made on the face of the receptacle (40) in
engagement with the refractory material (18). Nuts (104) are provided on these
ends. The bolts (100) are also rigidly connected to a plate (106) located at
the
back of the other part of the receptacle (40). This rigid connection is
achieved, for
CA 02334063 2001-02-02
13
instance, by welding the bolts (100) to the plate (106). The plate (106) is
preferably made of stainless steel. A compression spring (108) is coaxially
mounted around each bolt (100), between the interior wall of the cover (92)
and a
washer (110) resting against the plate (106) at the back of the receptacle
(40).
The bolts (100) act as guide rods to keep the probe (20) in registry with the
aperture (22) as the launder expands or contracts. The compression springs
(108)
provide a force which is constantly applied on the receptacle (40) to keep it
in
engagement against the side wall of the refractory material (18) even when the
frame (14) and refractory material (18) expand at different rates. It should
be
noted that the cover (92) should be designed to contain spillage of liquid
metal in
the event that the refractory material (18) breaks in the region of the
aperture (22).
Referring now to FIG. 4, the emitter coil (30) receives a signal from an AC
generator (120) in the form of a constant AC current. The AC generator (120)
is
controlled by a control module (122), consisting for example of a computer.
The
control module (122) is used to control the operation of the system (10) and
calculate the change in inductance into a value that is proportional to the
level of
liquid metal in the launder (12). The signal sent to emitter coil (30) is
preferably in
the form of a sinusoidal wave having a frequency between 0,1 and 10 kHz. The
preferred frequency is 1 kHz with a current of 500 mA. At the same time, the
control module (122) measures the inductance in the receiver coil (32) and the
temperature in the housing (40c) of the receptacle (40). The receiver coil
(32) and
the temperature sensor (42) are connected to corresponding analog-to-digital
converters (124,126), themselves connected to the control module (122).
The system (10) is preferably calibrated in two steps. In the first step, the
probe
(20) is calibrated by heating it in the controlled environment of a furnace.
The
signal values are recorded from room temperature to 400°C, for
instance. FIG. 9
shows a typical relationship between the signal and temperature of a 150 mm
probe. A second order equation of the relationship between signal and
CA 02334063 2001-02-02
14
temperature is calculated from the results and is downloaded in the non-
volatile
memory (122a) of the control module (122).
In the second step, the user enters the lower and higher signal that will be
measured by the system (10). It is done for instance by pressing an "Empty"
button (130) on a keyboard (132) when the launder (12) is empty. At this time,
the
lower signal value and the temperature are recorded in the non-volatile memory
(122a). The user then places a plate (not shown), which is larger that the
core
length, in the launder (12). The plate preferably has a thickness of at least
4 mm
and is made of the same metal or alloy to be transferred, for example a plate
of
aluminum if aluminum is used. Once the plate is in place, the user presses the
"Full" button (134) on the keyboard (132) and the higher signal and the
temperature are recorded in the non-volatile memory (122a). FIG. 10 shows a
typical relationship between the aluminum level and the relative signal of a
150
mm probe. The relationship is preferably described by a third order equation.
This
equation is downloaded in the non-volatile memory (122a) to be used during the
calculations.
In use, the system (10) generates and measures the inductance and makes
corrections for the change in temperature of the probe (20). It reads the
probe
signal and the temperature. The control module (122) compares the temperature
with the temperature used in calibration. The signal values are then corrected
according to the temperature equation. After temperature correction, the
system
(10) calculates the level using, for instance, the third order equation and
the lower
and higher values recorded by the user. The level is finally displayed or
recorded
in a display device (138). Furthermore, preset values can be provided to
trigger
alarm signals whenever the level reaches these values.
CA 02334063 2001-02-02
EXAMPLE OF CALIBRATION
A probe with a span of 150 mm was fixed on the side of the wall of a launder.
Two
equations have been downloaded into the non-volatile memory of the control
5 module. The first equation was the one relating to the signal with reference
to the
temperature. This first equation was expressed as:
S(T) _ (0,25 x T ) + (1,17 10 -3 x T 2) - 7,4 (1 )
10 where T is the temperature in °C and S(T) is the signal correction
for a temperature.
The second equation was the one for the determination of the metal level. This
second equation was:
Metal level (mm) _ (5,9 x S (~,o) ) - [ 9,5 x 10-2 x (S (0,°))2] + [5,5
x 10-4 x (S (~,0))3 ]
15 (2)
When the probe was fixed on the launder, the user pressed the "EMPTY" button.
At that moment, the signal value and temperature were recorded in the non-
volatile memory. A value of 20 000 arbitrary units (V ( EMPTY) ) was recorded
at a
temperature of 40°C. The user then placed the aluminium calibration
plate in the
launder and pressed the "FULL" button. A value of 21,000 arbitrary units was
recorded (V ( FULL) ).
During operation, a value of 21 551 arbitrary units and a temperature of
150°C
were measured. The processor of the control module then calculated the S (T)
from equation (1 ) at 40°C and 150°C, which gave values of 5 and
56 arbitrary
units, respectively. Then, the variation of signal relative to temperature,
called 0S
was given using the following equation:
OS = S ( 150°C) - S ( 40°C) (3)
CA 02334063 2001-02-02
16
0S was equal 51 arbitrary units. The processor had to determine whether the
operation temperature was higher or lower than the calibration temperature. If
the
operation temperature was higher than the calibration temperature, then the
equation (4) had to be used. If the operation temperature was lower than
calibration temperature, then the equation (5) had to be used. In the case,
equation (4) was used, giving a value for V (corr) of 21 500 arbitrary units.
This
value was then used for measuring the metal level.
V (corr) = V measured - ~S, for T calibration < T operation (4)
V (corr) = V measured + 0S, for T calibration > T operation (5)
The signal percentage S (~,o) was given using the following equation:
S (oia) _ [ (V (corr) - V ( EMPTY) ) ~ ( V ( FULL) - V ( EMPTY) )~ x 100 %
where V ( EMPTY) IS the value measured during the calibration EMPTY and V (
FULL)
is the value measured during the calibration full. In that case, S (°o)
equalled 50.
The metal level was calculated using this value in equation (2). That gave a
value
of 126 mm.
Of course, the foregoing equations were only given as examples and others can
be devised by a person skilled in the art.
EXAMPLE OF USE
The sensitivity and precision of a system (hereinafter "the system")
constructed in
accordance with the present invention were compared with that of a
commercially
available device which was based on the measurement of capacitance. In this
comparative test, the probe of the system was similar to that used in example
of
calibration, that is with a span of 150 mm. It featured an emitter coil and a
receiver
CA 02334063 2001-02-02
17
coil of 90 windings operating at the frequency of 1 kHz and with a current of
about
500 mA. The probe was calibrated by the procedure described earlier. It was
installed in the side of a launder through which was flowing an AA 5000 series
aluminum alloy. The temperature of the metal was 750°C, and the
temperature of
the frame of the launder was about 150°C.
The commercially available capacitance probe was also installed in the side of
the
launder, about 30 cm away from the probe of the system. Then, the performance
of the two probes was recorded on a same strip chart recorder over a period of
about 60 minutes. The resulting graphs are shown in FIG. 11, where they are
set
in superposed manner for comparison. The graph A relates to the prior art
device,
while graph B relates to the novel system. As can be appreciated, the signal
to
noise ratio of the system (B) is much better than that of the other device
(A). This
is clearly visible from the fact that the amplitude of the oscillations of the
recorder
trace in graph A are about four times wider than that those of graph B.
The above-described example shows that the system is sufficiently sensitive to
sense even a small increase in the level of the metal during the one hour
test. It
would be also able to sense a gradual increase followed by a slow decrease in
the
metal level. This was unlikely to be notice with the prior art device because
of its
higher noise to signal ratio, and thus its inherent lower sensitivity.
Although possible embodiments of the present invention have been described in
detail herein and illustrated in the accompanying figures, it is to be
understood that
the various aspects of the present invention are not limited to these precise
embodiments and that various changes and modifications may be effected therein
without departing from the scope or spirit of the present invention.