Note: Descriptions are shown in the official language in which they were submitted.
lZ~3336
Background of th _ nvention
A. Field of the Invention
This invention relates to the field of art of level
indicator control systems using capacitive probes for indication
and control of a substance in a tank.
B. Background Art
It is well known that the level of a substance, i.e.
a fluid or granular solid, in an open or closed tank or vessel
can be measured and controlled by many, fundamentally similar,
methods. Measurement and control is usually based on the concept
that the change in fluid level in the tank is equivalent to dis-
placing the top surface of the fluid.
In an earlier method of measurement and control,
floats were used to detect and regulate the fluid level in a
container. The method employs direct-actuated types of liquid
level detectors and is applicable to open tanks or vessels which
are subject to atmospheric pressure. However, when using closed
tanks, water level is detected in a system under pressure. An
arrangement used for this purpose includes one valve positioned
at the lowest fluid level point in the tank. Periodic opening
of these valves will establish the presence of either steam or
water at each valve permitting an inference to be drawn concern-
ing the actual water level in the tank.
Later detectors measured the fluid level in a tank by
sensing the hydrostatic head of the fluid and converting this
`~ `7~
833~
pressure measurement to actual fluid level height or fluid
volume. or open tanks, a pressure-gauge-type instrurnent may
be used. A connection is made to the pressure gauge at the
minimum or zero fluid level. The full scale range of the
gauge is made equal to the head of the fluid in the tank.
There are numerous variations of this method including adap-
tations to tanks where the pressure gauge cannot be located
at the zero level and where the medium to be measured is a
solid.
A pressure gauge is not practical for measuring fluid
level in a pressurized tank, since the actual level to be
measured represents only a very small equivalent percentage of
the static pressure of the fluid in the tank. Also, an added
difficulty is that unless the tank pressure is held constant,
the pressure gauge reading is of no value since the change in
pressure alters the initial zero level reading. To overcome
these problems, differential pressure measuring devices were
used to measure the fluid level in pressurized tanks. Connec-
tions are made at both high and low fluid levels, one to each
side of the di.fferential pressure device, i.e. a Bellows-type
meter. The separate connection to each side of the differential
pressure device provides for a balancing of the effect of
static pressure since it exerts the same force on both the high-
pressure and low-pressure side of the tank. Therefore, the
pressure head which actuates the detector is the difference
between the constant reference and variable fluid level in the
tank.
~2~ 36
Other devices, improvements and adaptations of float-
operated level sensors were developed based on the principles
previously discussed. For example, a float may be connected to
an electrical switch for providing an alarm, operating a sole-
noid valve or indicating when a discrete amount of fluid hasbeen poured in or removed from the tank. Floats may be used
to operate control valves directly to prevent further fluid
flow to the tank,~nd displacement-type float units may be used
to operate control units and remote transmitters.
Another method for detecting fluid levels in tanks
utilizes the concept that certain fluids will conduct electricity,
while air in a relative sense does not, so that the fluid level
may be established through the physical contact of the probe and
the conductive fluid. Since the change in fluid level is equi-
valent to displacing the top surface of the fluid, the usually
linear displacement may be measured by resistive, capacitive,
magnetic, or photoelectric transducers. Further methods of level
detection include temperature-sensing transducers, multi-turn
potentiometers operated by a float actuated cable and ultrasonic
and gama-ray adsorption.
A compu~er-based control system uses well known signal
acquisition input instrumentation to obtain analog signals from
sensors and transducers, such as capacitance probes, in the tank
and transmits them to the computer. To close the fluid level
control loop, D/A converters and digital output channels may be
833f~
used to transmit the signals used to drive on-off fluid level
controllers and actuators. Devices such as relays or stepper
motors for opening and closing pneumatic fluid valves are also
provided control signals from the computer along digital output
channels for controlling the fluid flow into and out of the tank.
The processor may, for example, compare the input signals from
the fluid level transducers with upper and lower set point
limits in order to control, in on-off, proportional, integral
or differential modes, the fluid flow to the tank to maintain
the desired liquid level within a predetermined range. Alarm
monitoring and faulty transducer detection can also be performed
by the computer.
Analog controllers may be used without a computer pro-
cessor for controlling the level of fluid in a tank. The analog
controller may either use its own set point reference voltage
to control fluid input to the tank or it may accept fluid level
set point li.mits from a central processor for the same purpose.
Output devices such as strip chart recorders using
properly scaled paper, calibrated meters with d'Arsonval move-
ment and digital displays have all been used to show the amount
and the height of fluid in tanks.
Summary of the Invention
An on-line level indicator control system is used for
automatically calibrating high and low set point levels of sub-
~'~Q~3~3~
stances such as fluids or solids in shaped containers. A
transmitter produces a present level signal that is proportional
to the capacitance between a probe positioned within a container
and the container. Logic means produces a time interval signal
that is a function of the duration of the present level signal
and represents the present level of the substance in the con-
tainer. Calibration switches provide the memory with calibrated
set points representing the percent of fullness of the container
at predetermined low and high fluid or solid levels as a func-
tion of the time interval signal. An extrapolator receives thehigh and low set points for providing calibrated range of fluid
or solid levels for the container as measured from the bottom
to the top of the container. A fail-safe mode controls the
filling or draining of the container when a short or open is
detected.
Therefore, it is an object of this invention to provide
on-line substantially automatic calibration of high and low fluid
level set points as a function of probe capacitance each corres-
ponding to a percentage of the tank that contains fluid at any
two identified high and low levels substantially within the range
from the bottom to the top of the tank.
It is another object of this invention to use the
arbitrarily selected high and low fluid level set points to pro-
duce a scale that includes substantially all of the possible
~P8313Çj
levels of fluid that could be in the tank in terms of percent
of tank filled as the actual fluid height is monitored through
the operation of the level indicator control system.
It is a further object of this invention to provide,
simultaneously, the actual level and amount(volume) of fluid in
the tank during the course of operation of the level indicator
control system.
It is an additional object of this invention to avoid
recalibration of the high/low set point levels during level
control operation using the same tank.
It is another object of this invention to eliminate
adjustments to the probe transmitter while in operation in the
field.
Still another object of this invention is to provide
a fail-safe mode when the line between the probe transmitter and
the control system opens or shorts.
Brief Description of the Drawings
Fig. 1 is a circuit diagram of the on-line level indi-
cator control system of the invention;
Fig. 2 shows the output waveform of the multivibrator
illustrated in Fig. 1.
Fig. 3 shows the output pulses of the cyrstal coupled
oscillator illustrated in Fig. L.
Fig. 4 shows the output signals of the AND gate illus-
~Z~83;36
trated in Fig. 1.
Fig. 5A-B illustrates a flow chart of the computer
program that detects when the system shown in Fig. 1 is in a
fail-safe mode.
~LZ~ 3~
Description of the Preferred Emboclimen
As shown in Fig. 1 level indicator control system 10
includes capacitance probe 12, transmitter 15, crystal coupled
oscillator 16, AND gate 34, counter 18, a control system comprising
microprocessor 20, manual data input and set point calibration
switches 44 and flow control circuit 22.
Probe 12 is positioned within a shaped container such
as tank 24. The probe, which may be substantially in the shape
of a cylinder or plate, forms a capacitor with tank wall 26. As
the substance 28, i.e., fluid or granular solid, fills tank 24,
the capacitance between probe 12 and wall 26 changes since the
varying amount of substance alters the dielectric properties of
the space between one capacitor element, the probe, and the
second capacitor element, the tank wall. Over time, the amount
of substance that is initially in the tank will not remain constant
and, therefore, the capacitance between the probe and the wall,
the probe capacitance, will vary as the level of substance in the
tank varies. Probe 12 produces an input signal along line 30 to
transmitter 15 which is a function of the change in probe capa-
citance. Probe capacitance is a function of the amount of probe
submerged by the substance or the distance between a plate type
probe and the surface of the substance below it. Transmitter 15
includes multivibrator 14 which changes its output frequency as
a function of the change in probe capacitance provided along line
30. The transmitter may be mounted in the head of the probe.
_9_
~Z~l33;36
Referring now to Fig 2, the signal transmitted frora
probe 12 to multivibrator 14 controls the width of multivibrator
output signal 54 along line 36. Specifically, the width of
waveform 50 of signal 54 is a function of the probe capacitance.
As the capacitance increases the output of multivibrator 14
produces waveform 52 which is wider than waveform 50 as shown
in Fig. 2 Output signal 54 of multivibrator 14 is applied
to logical AND gate 34 along line 3~. crystal coupled oscillator
16 provides a continuous series of pulses 56 as shown in Fig. 3
to AND gate 34 along line 39. The output of AND gate 34 is
provided along line 38 to counter 18 where only those oscillator
pulses, 57a, 57b, that occurred within the width of waveform 50
aid 52 respectively, are counted to provide a time interval
signal representing each of the widths of the waveforms of
multivibrator output signal 54. The number of oscillator pulses,
57a, 57b, counted within the width of each multivibrator output
waveform 50 and 52, as shown in Fig. 4, is a function of the
probe capacitance that substantially produced each waveform. This
capacitance of probe 12 is a function of the change in dielectric
properties of the substance 28 between probe and tank wall which
is also a function of the amount of substance in the tank or the
level of the substance in the tank. The oscillator, AND gate
and counter form a logic means for providing a time interval
signal to memory 42 in microprocessor 20.
In this embodiment, the time interval signal, which is
a function of the duration or width of each output waveform of
the multivibrator, is calibrated to represent the percentage of
-10 -
3~
the tank that contains fluid. Specifically, an output signal
waveform has a time interval, Tl, that is proportional to the
probe capacitance and represents, for this example, that 35/O of
the tank is filled with the fluid or granular soLid. Another,
longer waveform time intervalS ~2' may indicate that the tank
is 45% filled. Furthermore, slnce each time interval signal
represents the percentage of the tank that contains fluid and
since the contents of the tank will be filling and emptying
due to the conditions imposed on its use, it becomes important
0 to know, in certain situations, how high the fluid level is
pe_rC~ ~tQ e
,J''~`'' at any given Moment in terms of the pEec~ntag~of the tank
that contains fluid as well as whether the fluid has exceeded
a particular level or not. To accomplish this task, two known
time interval signals are provided, one that represents the
point at which the tank is substantially filled with fluid, a
high set point limit, and one that represents the point at which
the tank is substantially empty of fluid, a `low set point limit.
A scale of percentages about these two limits is produced where
the scale is calibrated to represent the percent fullness of the
tank at any given time about the limits set by the two high and
low set points. Each scale has elements that represent the full-
ness of the container as measured from the bottom of the container
to a point on the container identified by a scale element. The
method and device for measuring the level of fluid in a tank,
for controlling the amount of fluid in the tank and for providing
a scale that is calibrated in terms of percent of tank filled
and where each value of the scale is a function of the probe
lZ~83~3~
capacitance is as follows.
Probe 12 may be installed in an angular position suit-
able for the size and shape of any tank. Probes installed from
the side of the tank should be angled downward to allow the fluid
or granular solid to drain or slide off the probe. An adjustable
time delay may be included to eliminate spurious operation of
system 10 due to splashing or agitation of the fluid in the tank.
on Fig. 1, probe 12 is positioned vertically in cylindrically
shaped tank 24. Changing the level condition of substance 28 of
tank 24 will. change the capacitance between the probe and the
tank wall. Since this capacitance is transmitted to multivibrator
14 by way of a signal along line 30 and since the frequency with
which the multivibrator jumps between positive and negative volt-
age states is controlled by the capacitance signal transmitted
to it, the shape of output signal 54 from multivibrator 14 will
vary as a function of the probe capacitance.
Crystal coupled oscillator 16 provides clock pulses
along line 39 to gate 34. The multivibrator output signal wave-
forms are also provided as input to gate 34 along line 36. Gate
3~; produces, by logical conjunction of the clock pulses with the
multivibrator output waveforms, gated output signals along line
38 which are provided to counter 18. Counter 18 produces sig-
nals that substantially represent the width of each of the wave-
forms produced by the multivibrator. The digital signals pro-
duced by the counter are provided to a predetermined locationin microprocessor memory 42 along line 40. The counter output
~Z~8336
may represent a desired low or high fluid level set point
signal, representing the percentage of the tank that contains
fluid at substantially that moment, which is to remain in memory
indefinitely. Manual data input and set point calibration
switches 44 are provided for the purpose of entering relative
input parameters and storing the low and high fluid level
signals in memory 42 of microprocessor 20. The calibration
switches provide for fluid level signals to be stored in
memory that represent the operation measurement of substan-
tially any liquid level in the tank in terms of percent of
fullness of the tank on a scale of from 0% to 100%. Non-linear
cGnversion of the level signals to units of volume or flow
rate in open channel flow systems may be obtained through
suitable, well-known microprocessor based linearizing means
regardless of the shape of the tank. The low fluid level set
point signal may no necessarily represent the fluid level in
the tank when it is completely empty and the high fluid level
signal may not necessarily represent the fluid level in the
tank when it is completely full. The set points are auto-
matically calibrated in terms of the percent of fluid in tank
28. The high and low set points in memory 42 are provided to
I- extrapolator 46 along data lines 48~.Extrapolator 46 uses both
the high level set point, which may represent the tank fluid
level other than at 100% full, and low level set point, which
336
may represent the tank fluid level at a point other than when
the tank is empty. Extrapolator 46 will then produce the pro-
per range of percentages of fullness of the tank in the range
between empty (0%) and full (lQ0%) after operating on the two
low and high set points obtained from memory 42.
Extrapolator 46 in microprocessor 20 not only produces
a scale having the range of fluid levels in the tank in terms
of the percentage of fu].lness of the tank but may also provide
the volume of fluid in the tank at virtually any moment during
its use. Fluid flow rate may also be determined in open channel
systems. Extrapolator 46 may, by way of a table look-up feature,
determine fluid levels in oddly shaped containers and compensate
for non-linearities of probe configurations due to tanks of
different geometries.
System l provides a level control mode of operation by
comparing in logic and controller section 48 the actual fluid
level or volume in the tank with either the higher or lower set
point values provided to the section along line 51 from memory
42. For example, if overfill.ing of a tank is to be avoided and
the actual fluid level exceeds the high limit set point in memory
42, then a signal is provided from controller 48 along line 50
to proportional control valve 22, which may also include a stepper
motor or a similarly functioning device, to stop the flow of
fluid through line 29 into tank 28. Control relays may also be
~f2S~ 3~
used to control fluid flow into and out of the tank. If, on
the other hand, an empty condition cannot be tolerated and the
actual fluid level is below low limit set point in memory 42,
a signal :is provided along line 50 to proportional control
valve 22 to allow more fluid to flow into tank.
System 10 also provides a failsafe mode of operation
which provides a desired output, on display 60 for example, in
the event of a system failure such as a power or equipment fail-
ure. In this case, the fail-safe mode becomes operable when the
circuit shorts or opens along line 36. In a high fail-safe mode
a high level (unsafe) condition will be simulated for the
system. In the low fail-safe mode, the situation is reversed.
In the case of a short or open circuit, the low fail-safe mode
is implemented. If overfilling of the tank is to be avoided,
the high fail-safe mode will be used. If an empty condition
cannot be tolerated, a low fail-safe mode is required. As a
result, filling and draining of the tank will be controlled in
either mode. A computer program in microprocessor 20 is used
to detect when system 10 is in a fail-safe mode and responds
accordingly.
The functions of the program are basically shown
in flow chart 100 of Figs. 5A-B. Flow chart 100 describes
the sequence of events that occur during the execution of
the program that has been written in
-15-
;~
8~36
machine language for the 1802 microprocessor. The program
first initialized memory locations by defining them and pro-
viding them with initial values. This is performed at the pro-
gram initializatlon step 102. The program must first determine
what mode it is operating in and this is examined at decision
diamond 104.
There are a number of mode selections that can be made
in the on/off mode but the outcome is fundamentally the same
for all possible selections. Decision diamond 106 first requires
l.0 a determination as to whether calibration must be performed. If
calibration is not required, then the new input value will be
compared to the prior set point values for determining whether
system 10 should be or. or off. This occurs at block 108. In
block 110 the program will determine the high and low set point
values depending upon the mode selection made. The level status
of the present input valve will be made with relation to those
high and low set point values. At block 112 the fail-safe
switches are examined to determine the proper settings for each
switch. The program then returns to the program initialization
block 102. If one of the possible selections was made that
brings the program into the continuous mode at decision diamond
104, then another sequence of events will occur.
In decision diamond 114 a determination of whether or
not to calibrate will be made. If calibration is not required,
then block 120 shows that the span between the high and low levels
-16-
will be determined. In block 120a there is a determination
of the display value from the current count and span. Then
block 122 disylay value will be d:isplayed after using a look-
up table to obtain the scale. The analog representation of
the display value is provi.ded in block 124. The high and low
set point values can now be switched into computer memory 42
for further continuous mode operation. Switching the low and
high set poinL values in the computer memory takes place at
block 126. Branching now occurs from this point in the program
back to the first step of the sequence provided in block 108.
If calibration is required in either the on/off mode
or the continuous mode, the sequence of events are essentially
identical. First, a decision is made at decision diamonds 128a
and 128b to determine if calibration can continue. If calibra-
tion continues then a time interval signal will be produced
that is proportional to the capacitance between probe 12 and
container 26. The time interval signal which is a function of
the total number of oscillator pulses within the duration of the
signals provided by multivibrator 14, is obtained by the program
as indicated by blocks 130a and 130b. With this data the program
continues to perform either on/off calibration routines or con-
tinuous calibration routines at blocks 132a and 132b respectively.
These calibration routines are continued until branching occurs
to other portions of the program.
A test mode may also be selected that allows the program
336
to test display 60 and other analog outputs and sensors as
shown in block 134. As shown in block 136, if a sensor fails,
for example display 60 shows a zero which suggests an empty
container and analog outputs are provided at 4 milliamps which
is the lowest value of a 4-20 milliamp range. Display 60 shows
the same information when calibration fails to occur but a light
is also provided for this condition. This is performed in the
program as shown in block 136 of the flow chart.
Display device 60 may be used to accept a digital sig-
nal al.ong line 59 within a range of from 0 to 100%, for example,
to show the analog representation of the fullness of the tank
in terms of percent fullness based on the range of the scale
provided in extrapolator 46.
Once the probe and transmitter are installed in a
tank and the two high/low set points stored in computer memory,
the level control system will operate without further probe ad-
justment. This is especially important when the probe and trans-
mitter are used in explosion proof installations. The probe/
transmitter is, itself, ehplosion proof due, primarily, to its
low energy use. Calibrations are required as the system is moved
from one shaped tank to a different shaped tank although not
when reusing the same tank.
