Note: Descriptions are shown in the official language in which they were submitted.
AUTO-CALIBRATION
OF A SEEDER USING TANK SCALES WITH AUTOMATIC RATE ALARM
FIELD OF THE INVENTION
This invention relates generally to a material distribution apparatus and
method
which auto-calibrates the feed rate of material being distributed over an
area, and, more
particularly, relates to a seed planter for planting seeds over a planting
area. The apparatus
also provides an automatic rate alarm to the user when the feed rate is
outside a desired
range. The apparatus and method have been developed to allow for more accurate
metering
of material to ensure accurate coverage and to minimize wastage of material.
BACKGROUND OF THE INVENTION
Accurately applying a material such as seed or fertilizer to an area is an
important
aspect of farming. The material to be applied may be expensive and
distribution area is
often very large so mistakes in the rate of application can be expensive.
Prior art systems
for distributing material have been developed that allow an operator to vary
the rate of
application, however, there are no "on the fly" systems for automatically
calibrating the
apparatus while in use.
US Patent No. 7,765,944 discloses a system and method that employs a loadcell
connected to a seed container to determine the change in weight of product
over a measured
area to determine the actual seeding rate. This allows the operator to
determine if he needs
to adjust the actual seeding rate to the target seeding rate, but there is no
mechanism for
automatically adjusting the actual rate to the target rate during use.
US Patent No. 8,695,396 discloses a method of calibrating a distribution
apparatus
operation in which the apparatus captures the number of revs of the product
meter and
CA 3007174 2018-05-31
- 2 -
determines the change in quantity of product in the tank over an area. Using
this change in
quantity of product and the number of revs permits a meter flow quantity per
revolution of
the meter to be calculated so that the meter can be set to a desired
calibration setting based
on the calculated meter flow quantity per revolution of the meter, but there
is no automatic
calibration of the apparatus.
In view of the above discussion, there is a need for an improved distribution
apparatus and method for delivering material to a distribution area that
permits auto-
calibration of the feed rate of material to permit more accurate metering, and
an automatic
rate alarm if the feed rate is outside a desired range.
In the following detailed description, a specific application of the apparatus
and
method in the form of a seed planter and planting method is discussed in
detail by way of
example. It will be understood that the auto-calibrating material distribution
apparatus and
method described herein finds application beyond the agricultural field and
may be
employed in any environment where it is desired to accurately distribute
material over an
area.
SUMMARY OF THE INVENTION
Accordingly, there is described a distribution apparatus for delivering
material to a
distribution area comprising:
a frame supported for movement over the distribution area;
at least one tank for containing the material to be delivered to the
distribution area;
a load sensor to monitor the weight of the at least one tank and the material
loaded therein;
CA 3007174 2018-05-31
- 3 -
a distribution system to receive the material from the at least one tank and
distribute the
material in a distribution operation, the distribution system including a
metering system to
control the rate of material flow according to a target rate;
a control unit for receiving data from the load sensor with respect to the
weight of the at
least one tank and the material loaded therein during the distribution
operation, the control
unit recording data reported by the load sensor, processing the data and
comparing a load
sensor derived change in weight (AWL) with a theoretical change in weight
(AWT) based on
a calibration factor for the metering system to determine if the metering
system is operating
at the target rate, and assigning a revised calibration factor for the
metering system if the
target rate is outside a predetermined set point.
There is also described a distribution apparatus for delivering material to a
distribution area comprising:
a frame supported for movement over the distribution area;
at least one tank for containing the material to be delivered to the
distribution area;
at least one load sensor to monitor the weight of the at least one tank and
the material loaded
therein;
a distribution system to receive the material from the at least one tank and
distribute the
material in a distribution operation, the distribution system including a
metering system to
control the rate of material flow according to a target rate;
a control unit for receiving data from the load sensor with respect to a
change in weight of
the at least one tank and the material loaded therein during the distribution
operation, the
control unit processing the data and comparing a load sensor derived change in
weight
(AWL) with a theoretical change in weight (AWT) based on the target rate of
the metering
CA 3007174 2018-05-31
- 4 -
system to determine if the metering system is operating at the target rate,
and assigning a
revised calibration factor for the metering system if the target rate is
outside a predetermined
set point.
Optionally, a rate alarm may be activated if the calculated metering rate is
outside a
predetermined metering rate error.
In a further aspect, there is described a method of distributing material over
a
distribution area with a distribution apparatus including a frame supported
for movement
over the distribution area, at least one tank for containing the material to
be delivered to the
distribution area, at least one load sensor to measure the weight of the at
least one tank and
the material loaded therein, a distribution system to distribute the material
in a distribution
operation, the distribution system including a metering system to control the
rate of material
flow, and a control unit, the method comprising:
setting the metering system to a target rate;
distributing the material over the distribution area by operating the
distribution
apparatus and the distribution system over the distribution area;
monitoring the change in weight of the at least one tank and the material
loaded
therein by receiving data from the load sensor at the control unit during the
distributing step;
processing the load sensor data to determine a load sensor derived change in
weight
(AWL);
calculating a theoretical change in weight (AWT) based on a calibration factor
for the
metering system;
CA 3007174 2018-05-31
- 5 -
comparing the load sensor derived change in weight with the theoretical change
in
weight;
determining if the metering system is operating at the target rate; and
setting the metering system to a revised calibration factor if the target rate
is outside
a predetermined set point.
BRIEF DESCRIPTION OF THE DRAWINGS
Aspects of the present invention are illustrated, merely by way of example, in
the
accompanying drawings in which:
Figure 1 is a side elevation view of a distribution apparatus according to an
embodiment of the present invention;
Figure 2 is a cross sectional side elevation view of the tank with material
showing an
exemplary arrangement for monitoring the weight of the combination;
Figure 3 is a detail schematic view of the distribution apparatus of Figure 1
showing
sensors and their communication with the control unit of the apparatus;
Figure 4 is a graph showing the manner in which the measured data is
conditioned in
the control unit according to embodiments of the apparatus and method;
Figure 5a is a graph showing the manner in which the weight of the tank with
material is measured and processed for a "low accuracy" calculation according
to
embodiments of the apparatus and method
CA 3007174 2018-05-31
- 6 -
Figure 5b is a graph showing the manner in which the weight of the tank with
material is measured and processed for a "high accuracy" calculation according
to
embodiments of the apparatus and method
Figure Sc is a graph showing the derived weight error over time; and
Figure 6 is a schematic view of an embodiment of a loadcell calibration
apparatus for
use with the apparatus.
DESCRIPTION OF THE PREFERRED EMBODIMENTS
Referring to Figure 1, there is shown a distribution apparatus for delivering
material
to a distribution area in the form of a seed planter 10, commonly referred to
as an air seeder.
Planter 10 includes a main frame 16 supported for movement over the
distribution area.
Attached to the frame is at least one tank 12 for containing the material to
be delivered to the
distribution area. In Figure 1, there are shown two tanks 12 and 14 and other
tank
arrangements are possible. The tanks may be any suitable container for storing
the material
to be distributed. They may be hoppers, bins, boxes or other containers for
retaining the
material. Frame 16 and tanks 12 and 14 are supported for rolling movement over
the surface
of a distribution area by one or more wheels 18. In practice, frame 16 is
releasably attached
to a towing vehicle (not shown) by hitch 20 so that the planter 10 can be
towed over the
distribution area. A ground-engaging implement 24 for actually delivering the
material
below the surface of the distribution area includes a secondary frame 26
supported by
ground wheels 28 and connected to the rear of the main frame 16 by a secondary
hitch 30.
Alternatively, the ground-engaging implement may be positioned in front of the
air seeder or
the air seeder and the ground engaging implement may be formed on a common
frame.
Material from the tanks is delivered to the distribution area by a
distribution system
designed to receive the material from the tanks and distribute the material in
a distribution
operation to ground-engaging implement 24. In the illustrated embodiment, the
distribution
CA 3007174 2018-05-31
- 7 -
system comprises an air distribution system 34 which includes a fan 36
connected to a
material delivery conduit 38. Fan 36 directs air through the conduit 38 and a
material
metering system 40 is located at the lower end of each tank 12 and 14.
Metering system 40
controls the rate of material flow from the tanks according to a target rate
into the material
delivery conduit 38. The particular type of metering system is not important
to the
apparatus, however, commonly, the meter will be a volumetric meter such as an
auger or
fluted roller. The delivery conduit 38 consists of one or more individual
conduits beneath
each metering system with a product passage directing product into each
conduit. Each
conduit carries material in the air stream to a primary distribution manifold
50 which serves
to divide the flow of material into a number of secondary distribution lines
58. Each
secondary distribution line 58 delivers material to a secondary distribution
manifold 51
which serves to divide the flow of material into a number of tertiary
distribution lines 52.
Each tertiary distribution line 52 delivers material to a furrow formed by one
of a plurality
of openers 60 attached to the secondary frame 26 at transversely spaced
locations.
Optionally, a trailing firming or closing wheel 62 associated with each opener
60 firms the
soil over the material deposited in the furrow. Applicant's co-pending US
patent application
no. 15/710,633 filed Sept. 20, 2017 discloses an exemplary air seeder of the
type generally
discussed above, particularly with respect to ground engaging implement 24.
The air seeder described and illustrated in Figure 1 is an example of a Class
A
distribution system. An alternative embodiment using a Class B distribution
system could
also be used where the metering system feeds material into a plurality of
primary conduits
beneath each meter, where each primary conduit then carries the product to a
secondary
distribution manifold. An example of such a Class B distribution system is
disclosed in US
Patent No. 6,213,698 to Landphair et al. Alternatively, a Class C distribution
system could
also be used, where there is a plurality of meters attached to each tank that
feed material into
corresponding individual conduits that carry the product directly to each
furrow separately.
An example of such a Class C distribution system is disclosed in US Patent
Application
Publication No. US2018/0098485 to Beaujot et al.
CA 3007174 2018-05-31
- 8 -
While an air seeder is described and illustrated in Figure 1, embodiments of
the
present apparatus and method may also be used to distribute other materials
such as
fertilizer. In addition, as well as distributing materials comprising granular
particulates,
such as seeds or fertilizer, it will be understood that the system and method
described are
also capable of handling liquid material.
In order to control the operation of planter 10, a load sensor to monitor the
weight of
the tanks and the material loaded therein is employed. A control unit 80 (see
Figure 3)
receives data from the load sensor to continuously measure the weight of the
tanks and the
material loaded therein. During a distribution operation, the control unit
receives data from
the load sensor thereby recording the change in weight of the tanks and the
material loaded
therein as material is dispensed therefrom and delivered to the distribution
area. The control
unit dynamically processes the collected data over two dispensing periods and
compares the
load sensor derived change in weight (AWL) over the period with a theoretical
change in
weight (AWT) over the same period. The theoretical change in weight may be
calculated
based on the target rate of the metering system 40 in units of weight/unit
area and the
distribution area covered. This allows the control unit 80 to determine if the
metering
system is operating at the target rate. In a feedback loop, the control unit
may automatically
assign a revised calibration factor to the metering unit 40, if the target
rate is outside a
predetermined set range, or prompt the user if they would like to update the
factor.
The control unit 80, based on load sensor derived data, may also compare the
difference between AWL and AWT, and generate an error signal or alarm if an
error greater
than a set range is determined by an algorithm. An exemplary algorithm may be
based on a
load sensor capacity rating, a rate, an amount of total material product
metered, and load
sensor accuracy.
Preferably, the control unit 80 includes an input means to allow an operator
to enter
control parameters and a display for displaying information to the operator.
In the event that
a new calibration factor for the metering system is assigned, the display may
prompt the
CA 3007174 2018-05-31
- 9 -
operator to accept the revised calibration factor. The display may also allow
an operator to
be notified by the control unit if there are repeated and/or large differences
between the load
sensor derived change in weight over time and the theoretical change in weight
based on the
metering system target rate to alert the operator to an operating condition
that the control
unit cannot control via the feedback loop. Such operating conditions may
include metering
problems that arise due to, for example, tank pressurization issues,
mechanical issues with
the metering system, and partial or full blockage of the distribution
material.
Referring to Figure 2, there is shown a preferred arrangement for using a load
sensor
to measure the combined weight of the tank 12 and product 13. The load sensor
comprises
at least one loadcell 102 positioned between tank 12 and the frame 16 and
configured to
detect the combined weight of the tank and the material therein. In practice,
two or more
loadcells would be positioned to provide a more accurate reading of the
weight. Depending
on the structure and configuration of the tank 12, it may be preferable to
mount the tank to a
sub-frame 16a and fit loadcells 102 between the sub-frame and main frame 16.
Turning to Figure 3, there is shown a detailed schematic view of the system of
Figure 1 showing various sensors that may be used in the collection of data.
It is
advantageous to include at least one accelerometer 70 for the entire apparatus
or, preferably,
an accelerometer with each loadcell 102 to measure acceleration to reduce the
magnitude of
the dynamic error in measuring the tank weight. This arrangement permits the
weight of the
tanks to be measured despite the entire distribution apparatus moving as it is
towed over the
surface of the distribution area. Similarly, it is preferable to include an
inclination sensor
72, such as a gyroscope or inclinometer, to measure the angle or inclination
of the tanks as a
group, or each tank individually, to the horizontal to take into account
tipping or rolling of
the tanks and material during a distribution operation for use in calculating
the static weight
value. Data from the various sensors is transmitted to control unit 80 where
data analysis
occurs. In the case where multiple loadcells 102 are associated with a tank
12, preferably, a
junction box 75 sums data from the associated loadcells 102 to communicate the
raw
loadcell data to the control unit 80. The inclinometer 72 and accelerometers
70 are used to
CA 3007174 2018-05-31
- 10 -
condition the raw loadcell data in preparation for averaging the data in order
to estimate the
weights under dynamic conditions.
Figure 4 is a graph illustrating how the raw loadcell data is conditioned.
Line 180
shows the raw weight data measured by the loadcells over time. Line 180 is non-
linear as
the collected weight data is affected by hill angle and the dynamic noise
created by the
motion of the apparatus over the ground. Hill angles result in the loadcells
measuring too
low a weight and dynamic noise results in additional large weight variations.
Line 200
shows the weight data partially corrected to remove the error created by hill
angles using the
inclinometer data and to reduce some of the dynamic noise using the
accelerometer(s) data.
The control unit 80 then averages this partially corrected weight data to
determine the
dynamic average loadcell derived weight as represented by line 202.
In general, there are two primary types of calculations processed at the
control unit
for embodiments of the present apparatus and method; firstly, a short term
relatively "low
accuracy" calculation, and, secondly, a long term relatively "high accuracy"
calculation. It
is the "low accuracy" calculation that permits quick alerts to large metering
rate errors to be
generated, whereas the "high accuracy" calculation permits the described
apparatus and
method to perform its automatic calibration function. Generally, the short
term relatively
"low accuracy" calculation will be comparing the change in weight over a
period of less
than fifteen minutes and providing accuracies of approximately 5% error or
larger. The long
term relatively "high accuracy" calculation will be comparing the change in
weight over a
period of greater than fifteen minutes and providing accuracies of
approximately 5% error or
smaller. These periods and accuracies are only generalities and it will be
appreciated that
they will vary depending on various factors which influence the calculation
including, but
not limited to, the weight of product metered out, dynamic noise, the system's
capability to
average out that noise, loadcell accuracy, and the loadcell capacity.
Referring to Figure 5a, a graph of tank weight over time, a detailed
description of the
"low accuracy calculation will be described. Line 200 represents the
conditioned dynamic
CA 3007174 2018-05-31
- 11 -
weight data collected over time by the control unit 80 as a distribution
operation is
performed and line 202 represents the dynamic average loadcell derived weight.
A first
averaged weight Wavg 1 is calculated by averaging the "pool" of conditioned
dynamic
weights measured over a period between time ti and time t2 with the average
weight having
occurred at the mid-point tml between ti and t2. A second averaged weight
Wavg2 is
calculated in the same manner over a period between time t2 and t3 at midpoint
tm2. The
difference between Wavg I and Wavg2 is then used to determine the loadcell
derived weight
change of the product AWL over the period between the time midpoints tml and
tm2. This
loadcell derived change in weight AWL is compared to the theoretical change in
weight AWT
based on the target rate of the metering system over the same period to
determine if the
metering system is operating at the target rate.
For the "low accuracy" calculation the period between tml and tm2 is
relatively
short. It is proposed this period will be between approximately two and
fifteen minutes in
length. The "pool" of conditioned dynamic weight in this "low accuracy"
calculation is
much smaller than the "pool" of conditioned dynamic weight in the "high
accuracy"
calculation, so any rate errors create a larger effect on the pool allowing
for quicker alarms
to metering rate errors but of lower accuracy. Also, for the "low accuracy"
calculation the
length of time between the first averaged weight and the second averaged
weight will
remain constant and continuously update in a rolling fashion, that is, the
first averaged
weight calculation will roll forward at the same speed as the second averaged
weight
calculation. For example, a future calculation may be determining the
difference between
Wavg2 and Wavg3 to determine the loadcell derived weight change of the product
AWL
over the period between the time midpoints tm2 and tm3. It is contemplated
that there may
be value in having multiple rolling rate alarms, each with a different rolling
average time.
Each time span will have a different rate error % range, For example, a
calculation with a
short time span will have a lower accuracy, and, thus, will only be capable of
determining
larger metering rate errors, whereas a calculation with a long time span will
have a higher
accuracy and will be capable of determining smaller metering rate errors.
CA 3007174 2018-05-31
- 12 -
Referring to Figure 5b, a graph of tank weight over time, a detailed
description of the
"high accuracy" calculation will be described. Line 200 represents the
conditioned dynamic
weight data collected over time by the control unit 80 as a distribution
operation is
performed and line 202 represents the dynamic average loadcell derived weight.
A first
averaged weight Wavgl is calculated by averaging the "pool" of conditioned
dynamic
weights measured over a period between time ti and time t2 with the average
weight having
occurred at the mid-point tml between ti and t2. A second averaged weight
Wavg3 is
calculated in the same manner over a period between time t2 and t3 at midpoint
tm3. The
difference between Wavgl and Wavg3 is then used to determine the loadcell
derived weight
change of the product AWL over the period between the time midpoints tm I and
tm3. This
loadcell derived change in weight AWL is compared to the theoretical change in
weight AWT
based on the target rate of the metering system over the same period to
determine if the
metering system is operating at the target rate.
For the "high accuracy" calculation, the period between tml and tm3 is
relatively
long. It is proposed this period will be approximately greater than fifteen
minutes in length.
The greater the length of time between tm I and tm3 the larger the "pool" of
conditioned
dynamic weight and the greater the accuracy of the calculation. For the "high
accuracy"
calculation, this length of time between the first weight and the second
weight will continue
to increase along with the "pool" of data also increasing with the result that
the accuracy of
the calculation will also increase. This increasing accuracy is shown in
Figure 5c, a graph
illustrating loadcell derived weight error E versus time t. Line 206
represents the loadcell
derived weight error E and, as time increases, this error decreases
exponentially. For
example, referring again to Figure 5b, the second calculation for the "high
accuracy"
calculation may now involve determining the difference between Wavg2 and Wavg6
to
determine the loadcell derived weight change of the product AWL over the
period between
the time tm2 and tm6. The period for this second calculation between tm2 and
tm6 is
greater than the first calculation between tml and tm3, therefore, the
accuracy of this second
calculation will be greater than the first calculation. The rate error
magnitude required to
trigger an automatic rate alarm may start at a high value and decrease over
time, essentially
CA 3007174 2018-05-31
- 13 -
following the maximum potential accuracy of the calculation as illustrated by
line 206 in
Figure 5c. There may also be a user adjustable factor in the system to account
for sensitivity
preferences between users.
The "high accuracy" calculation may also continuously update in a rolling
fashion
similar to the "low accuracy" calculation, but with a much larger period
between the first
and second weight. For example, the control unit 80 may only be able to buffer
and process
a limited amount of data, and, once this capacity is reached, the first weight
calculation may
roll forward at the same speed as the second weight calculation. It is
contemplated that the
period between the first averaged weight and second averaged weight for the
"high
accuracy" calculation may be one hundred and fifty minutes, but this period
could be
smaller or larger depending on the control unit and desired accuracy.
In both the "low accuracy" and "high accuracy" calculations described above,
the
first weight of the product is determined by averaging the dynamic conditioned
loadcell
weights over a period during the distribution operation. Alternatively, and
preferably, the
first initial weight could be determined by using the static weight measured
while the unit is
stationary, thus eliminating the time required to average the dynamic
conditioned weights.
For example, referring again to Figure 5b, in the "high accuracy" calculation,
preferably, a
first weight W1 may be determined while the tank is stationary, for example,
soon after
filling the tank. A second averaged weight Wavgl is calculated by averaging
the "pool" of
conditioned dynamic weights measured over a period between time ti and time t2
with the
average weight having occurred at the mid-point tm 1 between ti and t2. The
difference
between W1 and Wavgl is then used to determine the loadcell derived weight
change of the
product AWL over the period between time t1 and tml, which is quicker than the
first
method described using the difference between Wavg 1 and Wavg3 over the period
between
time tml and tm3.
In both the "low accuracy" and "high accuracy" calculations described above,
the
dynamic conditioned loadcell weights are averaged over a period, however, it
is obvious to
CA 3007174 2018-05-31
- 14 -
one skilled in the art that instead of averaging over a period, the dynamic
conditioned
loadcell weights could be averaged over a first area seeded to obtain the
first averaged
weight, and, then, the same calculation could be performed for a second seeded
area to
obtain the second averaged weight. Alternatively, the system could calculate
the area or
time required to dispense a derived loadcell weight, and compare this to the
theoretical area
or time based on the target rate and area covered or time elapsed.
In addition, it is contemplated that the described distribution apparatus will
automatically perform stationary weight measurements of the tanks and material
and
associated calculations each time the apparatus comes to rest for longer than
a pre-
determined period to confirm the running average dynamic calculations.
Referring to Figure 5a, refilling of the tank occurs as shown at 204.
Any time material is added to or subtracted from the tank (for example,
filling or emptying a
tank) without accumulated acres (or metering system revs), the weight of such
material is
not added to or subtracted from, respectively, the loadcell derived change in
weight AWL
over the previous operating period or area. No averaging of weight data will
occur when
product is being added or removed from the tank without the accumulation of
acres or meter
revs. For both the "low accuracy" and "high accuracy" rate alarms, it is
preferable that the
initial weight and all the data points that will be used for the next
averaging calculation are
offset (either + or -) by the change in weight of the product (for example,
when filling or
emptying a tank) to eliminate the time required for the alarm to activate.
Alternatively, the
system may also be reset such that a new initial first weight is determined
either by this new
static weight or calculated by averaging the weight over a period. Any time
the metering
system calibration factor is changed, the running average comparison between
AWL
maintained by the control unit is preferably reset and the comparison starts
over.
In a preferred arrangement, the metering system 40 comprises a meter such as
an
auger or roller for feeding material from the tanks. In operation, a rotatable
element of the
meter is rotated with each revolution of the element dispensing a defined
weight of material
CA 3007174 2018-05-31
- 15 -
from the tanks based on a calibration factor of the meter measured in the
weight of material
dispensed per revolution of the meter, for example, pounds/revolution. After
leaving the
meter, the material is deposited in conduit 38 to be delivered by air to
ground engaging
implement 24 as explained above. When the apparatus is engaged in a
distribution
operation, the control unit 80 continuously calculates a calibration factor
(pounds/revolution) for the meter in accordance with the process described
above. If the
calibration factor moves outside a set range, the operator is prompted to
accept the new
calibration factor or the operator may be warned to stop, as there may be a
mechanical issue
causing repeated or large rate differences from the targeted rate.
In the case of the material to be distributed being a liquid, the metering
system may
include a liquid pump or an orifice/gate mechanism to deliver a controlled
amount of liquid
per cycle of the pump or per operation of the mechanism. For example, in one
embodiment,
the liquid pump may be a fixed displacement pump, and the speed of the pump is
adjusted to
change the rate of liquid delivery. In an alternative arrangement using a
constant speed
pump, the rate may be adjusted through the use of a regulator valve associated
with the
pump which acts to return unused flow back to tank. In such a configuration,
the calibration
factor adjusted to control liquid flow is based on the position of the
regulator flow. In a still
further arrangement, a flow meter or a pressure gauge is associated with the
liquid pump,
and the calibration factor of the metering system is based on the flow meter
and pressure
gauge reading.
Other metering system arrangements may also be used with the apparatus. In an
alternative arrangement, the metering system may comprise multiple meters such
as augers
or rollers associated with each one of a plurality of tanks for feeding
material from a
particular tank. These multiple meters can be driven from a common shaft so
they all turn at
the same rate. They may also be individually driven with their own respective
prime mover,
such as an electric motor or hydraulic motor. In operation, each meter is
rotated with each
revolution dispensing a defined weight of material from the tank based on a
calibration
factor of the meter measured in the weight of material per revolution of a
rotating element of
CA 3007174 2018-05-31
- 16 -
the meter for example, pounds/revolution. After leaving the meter, the
material is deposited
in a conduit to be delivered by air to a ground engaging implement as
explained above.
When the apparatus is engaged in a distribution operation, the control unit 80
continuously
calculates a calibration factor (pounds/revolution) for the meter assembly in
accordance with
the process described above. If the calibration factor moves outside a set
point, the operator
is prompted to accept the new calibration factor, or the operator may be
warned to stop as
there may be a mechanical issue causing repeated or large rate differences
from the targeted
rate. As there may be more than one metering system feeding material from
multiple tanks
any change in the calibration factor will be applied appropriately to each
metering system.
In a still further arrangement, the metering system may be separated from the
tanks,
and located in another preferred location such as the drill. In this
embodiment, the metering
system(s) may be fed with material from the tanks by the method of induction
as described
in US Patent No. 7,021,224 to Mayerle et al. Alternatively, instead of
induction, the
metering system(s) may be fed with material from the tanks by another
mechanical meter
such as an auger or roller distribution system. When the apparatus is engaged
in a
distribution operation, the control unit 80 continuously calculates a
calibration factor
(pounds/revolution) for the meter assembly in accordance with the process
described above.
Since the described embodiments of the distribution apparatus rely on load
sensors
accurately measuring the weight of the tanks and the material therein, it is
expected that a
calibration system for efficiently and reliably calibrating the load sensors
may be
incorporated into the apparatus.
For example, in an arrangement where the load sensor comprises at least one
loadcell
positioned between a tank and the frame, the calibration system may comprise
means for
applying an external force to each tank in the form of one or more telescoping
cylinders to
load the tank to a known load value. Figure 6 provides a schematic view of
such a
calibration arrangement positioned between tank 12 and frame 16. One or more
telescoping
calibration cylinder(s) 90 applies an external, independently measured force
via links 96
CA 3007174 2018-05-31
- 17 -
between frame 16 and cylinder 90 and links 98 between the cylinder and tank 12
to load the
tank to a known load value. Calibration cylinder(s) 90 may be hydraulic,
pneumatic, or
electric actuators. The load value applied by each calibration cylinder may be
measured
with a gauge indicating cylinder pressure (not shown) or with a calibration
loadcell or cells
92 positioned in series with each cylinder. Preferably, each tank may be
loaded by its
associated calibration cylinders to one or more load values to calibrate at
more than one
point. For example, a lower load value, an intermediate load value, and an
upper, near full
scale load value may be used. At each selected loading point, the loadcells
associated with
the tank are read to determine if the loadcells are accurately recording the
applied load based
on the gauge or loadcell associated with the calibration cylinders.
The above calibration approach offers significant savings in time and effort
compared to current methods which tend to rely on an operator weighing a truck
full of
product on existing scales, filling a tank of a distribution apparatus from
the truck, and then
weighing the truck again. The reduction in weight of the truck represents the
weight of
material delivered to the tank for checking against the increased weight of
the tank as
measured by the installed weight measuring system of the tank.
In addition, an automatic warning may also be provided to the operator if the
control
unit 80 detects that there may be an error with one or more of the tank
loadcells. For
example, the control unit 80 stores the weight of each tank when the tank is
empty, and,
therefore, can compare this known weight with the weight that should be
detected by each
loadcell when the tanks are empty. When a tank is empty or near empty based on
the
theoretical weight calculation, if the control unit 80 detects a weight on a
loadcell that is a
predetermined amount higher or lower than its expected weight, an alarm will
automatically
be provided to the operator to calibrate the loadcell. Alternatively, every
time the user resets
a tanks weight to zero, the control unit 80 can check how much weight is being
zeroed out
compared to the known empty tank weight and provide an alarm, if it is beyond
allowable
limits.
CA 3007174 2018-05-31
- 18 -
Although the present invention has been described in some detail by way of
example
for purposes of clarity and understanding, it will be apparent that certain
changes and
modifications may be practised within the scope of the appended claims.
CA 3007174 2018-05-31
