Europaisches Patentamt

(19) European Patent Office

Office europeen
peen des brevets
brevets EP0 631 115 B1

(12) E U R O P E A N PATENT S P E C I F I C A T I O N

(45) Date of publication and mention (51) IntCI e G01G 1 1 / 0 8

of the grant of the patent:
22.04.1998 Bulletin 1998/17

(21) Application number: 94108584.7

(22) Date of filing: 04.06.1994

(54) Loss-in-weight weighing with bulk density determination using level s e n s i n g

Differentialdosierwaage mit Messung der Schuttdichte durch Fullstandsbestimmung
Pesage differentiel avec mesurage de la densite par detection de niveau

(84) Designated Contracting States: (72) Inventor: Ricciardi, Ronald J.

BE CH DE FR GB LI NL Woodcliff Lakes, New Jersey 07675 (US)

(30) Priority: 25.06.1993 US 82993 (74) Representative: Schickedanz, Willi, Dipl.-lng.

Langener Strasse 68
(43) Date of publication of application: 63073 Offenbach (DE)
28.12.1994 Bulletin 1994/52
(56) References cited:
(73) Proprietor: ACRISON, INC. WO-A-89/07574 DE-A-4 014 322
Moonachie New Jersey 07074 (US) GB-A- 2 179 588 US-A- 4 320 855
US-A- 4 378 897

DO

CO
CO Note: Within nine months from the publication of the mention of the grant of the European patent, any person may give
notice to the European Patent Office of opposition to the European patent granted. Notice of opposition shall be filed in
o
a written reasoned statement. It shall not be deemed to have been filed until the opposition fee has been paid. (Art.
a.
99(1) European Patent Convention).
LU
Printed byJouve, 75001 PARIS(FR)
 EP0 631 115 B1

Description

FIELD OF THE INVENTION

5 The present invention relates to the field of material feeding systems, particularly, weight-loss type feeding systems
having a periodic refill requirement.

BACKGROUND OF THE INVENTION

10 Precise metering of liquids and dry solids, such as amorphous powders, is an important but difficult task in many
industrial applications and processes. Whenever such materials are continuously fed into a process, the feed rate at
which material is delivered to the process must be controlled, just as with any other material used as an ingredient or
additive in a continuous or batch process.
Types of material feeding systems include volumetric feeding systems and gravimetric feeding systems. As the
is name implies, volumetric feeding systems dispense material by volume. They employ a displacement mechanism of
some sort (for example, an auger mounted below a supply vessel and feeding a fixed volume of material per auger
rotation) operating at a set speed. This results in feeding a known volume of material. The weight of material fed can
generally be determined based on the material's bulk density. Volumetric systems cannot always be relied on to produce
good results because actual conditions, such as the density, flow and handling characteristics of the material, change
20 during the feeding process.
Where a high degree of accuracy is required, gravimetric feeding systems are employed. These systems monitor
their own feeding performance and automatically correct for variations from the desired or set feed rate regardless of
product characteristics. Gravimetric systems control the feeding of material by weight, thereby compensating for chang-
es in density and/or irregular product density, flow or handling characteristics. For this reason, these systems can
25 provide much higher accuracy than volumetric feeding systems.
For weigh feeding of materials, there are various types of gravimetric systems used. Two typical types are weigh-
belt and loss-in-weight feeding systems. Typical weigh-belt gravimetric feeding systems measure the weight of the
material passing across the weigh-belt during operation, that is, while the belt is dispensing material. This measured
weight is compared to an expected or "set" weight, causing the generation of a control signal. The control signal either
30 increases, decreases or maintains the speed of the motor driving the belt to achieve a desired feed rate. While such
weigh-belt systems offer good accuracy for many uses, material sometimes accumulates on the belt or other critical
components and thereby causes the control system to improperly adjust the feed rate. The problem is compounded if
the accumulated material builds up and then falls off critical sections of the weighing mechanism causing shifts in the
calibration of the scale's "zero" calibration setting. Shifts in the scale's "zero" calibration setting can also be caused by
35 temperature variations or differing tension of the weigh belt. Also, because of the flowing nature of certain materials,
it can be difficult to reliably keep the material on the belt, thus causing additional inaccuracy. Similar to a weigh-belt
system is a weigh auger system, which operates on the same principle as a weigh belt system, but uses an auger in
lieu of a belt for dispensing product.
The weigh-belt and weigh-auger types of gravimetric feeding systems, therefore, while using feedback principles
40 to offer control, have inherent limitations that can seriously impair the accuracy of these systems.
Another type of gravimetric feeding system usable for dry bulk solid and liquid materials is the loss-in-weight sys-
tem. In a loss-in-weight system, the gradual decrease in weight of the entire system (including the discharge or feeding
mechanism, such as an auger device or a pump, a supply vessel or dispensing bin attached to the feeding mechanism,
and the material itself) caused by the constant release of material from the scale-mounted feeding mechanism is
45 monitored and compared to an expected decrease in weight to assess performance. Unlike the weigh-belt and weigh-
auger gravimetric systems, loss-in-weight systems are not affected by material adhering to or releasing from critical
areas of the weighing mechanism, or a shift in the scale's "zero" or calibration, since the entire system is weighed
continuously.
Preferably, the loss-in-weight feeding system is designed with a counterbalance so that the feeding mechanism
so is tared-out on the scale such that the scale produces a zero weight signal when the feeding mechanism is empty.
Consequently, the scale only measures the weight of the material in the feeding mechanism. Knowing the size and
type of the supply vessel and the feeding mechanism, the amount of weight attributable to the material in the supply
vessel can be determined. In loss-in-weight systems, a control system continually (or with great frequency) monitors
a signal indicating the decreasing total weight of the material, and adjusts the feed rate output to maintain an accurate
55 rate of weight loss over time in relation to the selected feed rate (operator-input set rate). If a weight loss amount over
a given time period is larger than expected as sensed by the weighing system, the control system commands the
feeding mechanism to slow down. Similarly, if the weight detected becomes too high, indicating that the weight loss
has been less than expected, the control system orders an increase in the material output flow rate to cause the system

2
 EP0 631 115 B1

to catch up with the expected feed rate.

Representative control systems for loss-in-weight feeding systems are shown and described in United States
Patents Nos. Re. 30,967, Re. 32,101 and Re. 32,102, incorporated herein by reference, and all assigned to the assignee
of the present invention, Acrison, Inc. Also, loss-in-weight feeding systems sold by the assignee of the present invention,
5 Acrison, Inc., including microprocessor-based control systems therefor, are commercial examples of control systems
for loss-in-weight feeding systems.
Depending upon the desired accuracy of feed rate, volumetric, weigh-belt/weigh-auger or loss-in-weight systems
are used for feeding materials.
During standard operation, different samples of dry solid material can be of greatly varying bulk densities, even
10 though the samples consist of the same material. In existing loss-in-weight systems, the weight of the material in the
loss-in-weight system is continuously (or almost continuously) monitored. Automatic refill of the system's integral supply
vessel is initiated when the monitored weight is sufficiently low and continued until the monitored weight reaches a
predetermined high weight, at which point refill is stopped. If the material has a sufficiently lower bulk density than
initially determined, however, the material may not be heavy enough to trigger the shut off of the automatic refill mech-
15 anism, even when the supply vessel is full because the predetermined high shut-off weight was determined based on
the higher density (weight) material. Consequently, with a low bulk density material, refill continues, forcing the loss-
in-weight system to remain out of gravimetric control (thereby reducing accuracy) and, perhaps, damaging the feeding
system.
In other loss-in-weight systems, a material sensing probe is positioned in the supply vessel. When the probe senses
20 a predetermined level of material, refill is shut off. See, e.g., U.S. Patent Nos. 4,320,855, Re. 32,101, and Re. 32,102.
Loss-in-weight feeding systems generally have a control panel that displays the volume of material in its integral
supply hopper. A controller determines the volume to be displayed based, in part, on a stored bulk density value for
the material. When the bulk density of the actual material in the supply hopper differs from the bulk density value stored
by the controller, the displayed volume is less accurate than is desired.
25 A relatively accurate bulk density measurement is required to control the feed rate as well. During operation, there
are times when a loss-in-weight feeding system will not operate in loss-in-weight mode and will operate temporarily in
a volumetric mode. For example, during refill, the weight signal is actually increasing as material is being added to the
supply vessel and so the actual loss-in-weight signal cannot be used to determine the amount of material being fed.
The weight of the material fed is often determined instead based on the volume of material displaced by a displacement
30 mechanism (such as an auger) and by the density of the material. When the density employed by the controller is not
accurate, the determination of the amount of material fed during such a period of time is less accurate than is desired.
At times during operation, the feeding system does not use loss-in-weight to control but rather controls based on
an estimated motor speed. The controller selects a motor speed which corresponds to the set rate by interpolating
from the known feed rate for a given motor speed. This is called the "fast start" routine. For example, when the feeding
35 system is first started or when the operator inputs a new, different set rate, the system cannot operate in a loss-in-
weight mode because there are insufficient weight readings at the new set rate to accurately measure the material
flow and properly control the feeding system. Therefore, to obtain quickly an accurate feed rate at start up, the controller
determines a motor speed estimate corresponding to the operator-input set rate based on the "maximum feed rate."
Maximum feed rate (or "mfr") is the rate at which material flows when the motor is running at 100% of motor speed. If
40 the maximum feed rate is 100 Ibs/hr (45 kg/hr) at 100% of motor speed and the set rate is 50 Ibs/hr (23 kg/hr), the
controller directs the motor to run at 50% of motor speed. The maximum feed rate, however, changes with the density
of the material. Consequently, when the bulk density of the material used by the controller is not accurate, the feed
rate in "fast start" mode is less accurate than is desired.
Furthermore from Figure 1 of US-A-4378897 a volumetric feeding apparatus with a high level probe 20 and a low
45 level probe 21 is known.
It is an object of this invention to provide an improved method and device capable of calculating the density of the
material in the supply vessel.
It is a further object of this invention to provide an improved material feeding system capable of using the newly
calculated density of the material to improve control of the feed rate of the feeder.
so It is a further object of this invention to provide an improved material feeding system capable of accurately calcu-
lating and displaying the volume of material in the supply vessel.
It is a further object of this invention to provide an improved method of calculating the density of material located
in a supply vessel.
It is a further object of this invention to provide an improved method of improving the control of a material feeding
55 system based on the newly calculated density of the material.
This object is solved with the method of claim 1 and the device of claim 14.

3
 EP0 631 115 B1

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a schematic view of a feeding system of the present invention including a high level probe;
FIG. 2 is a block diagram showing the method of calculating material density after refill employed by the system
5 of FIG. 1; and
FIG. 3 is a block diagram of the method of maximum rate/bulk density auto compensation employed by the system
of FIG. 1.

DETAILED DESCRIPTION
10
FIG. 1 shows schematically a portion of a material feeding system 1 including a materials supply vessel 2, discharge
mechanism or feeding assembly 10 and controller 12. The materials supply vessel 2, such as a supply hopper 4,
contains a level of material 3. The supply vessel 2 is commercially available from Acrison, Inc., assignee of this appli-
cation, for example, model 400 Series feeders. The feeding assembly 10 is positioned beneath the supply vessel 2
is such that material 3 flows from the supply vessel 2 into the feeding assembly 10. In a dry solids system, the feeding
assembly 10 is typically an auger metering mechanism 14 with rotating auger 15 driven by a motor 13. When fluids
are being fed, the feeding assembly is typically a pump. Acceptable loss-in-weight metering mechanisms are, for ex-
ample, Acrison model 105, 140 or BDF feeder.
A refill supply source, such as refill hopper 20, is positioned above the supply vessel 2. The refill hopper 20 has a
20 top inlet aperture 22 and a bottom outlet aperture 23 and is also filled with material. A refill valve 21 is disposed at the
bottom aperture 23. Material to be fed is loaded into the refill hopper 20 through the inlet aperture 22.
A weighing mechanism or scale 11, for example, Acrison model 401 , 402, 404, 405 or 406, is located beneath the
feeding assembly 10. A high level probe 30 is positioned within the supply vessel 2, preferably at a high level L. The
high level probe 30 is capable of detecting material in the supply vessel 2 at the level of the high level probe 30 and
25 producing a signal to a weigh feeder controller 12 indicating that material is at that high level. The level L corresponds
to a known volume within the supply hopper 2 up to that level. While it is preferred that the high level probe 30 be
located near the top of the supply vessel 2, the probe 30 could be lower in the supply vessel 2 and still practice the
present invention. Acceptable probes can be point level indicators or capacitive proximity sensors. Such probes detect
material in various ways and are commercially available, for example, from Bindicator of Port Huron, Ml as the Roto-
30 Bin-Dicator or Breakdata 2200, from Custom Controls Co., Inc. of Broomall, PA as the Level Lance, and from Drexel-
brook Engineering Co. of Horsham, PA as the LCS Point Level Control Model 10. Acceptable capacitive proximity
sensors are commercially available from Pepperl & Fuchs, Inc., and others.
Preferably, the weigh feeder controller 12 is mounted in a control panel 31 having a panel display 32 such that the
controller 12 can receive operator input from a keypad 33 on the panel 31 . Particularly, the controller 12 preferably is
35 able to receive input regarding the selected feed rate (or set rate) from an operator and to command the panel display
32 to display the volume of material 3 in the supply vessel 2. One acceptable controller to interface with the high level
probe 30 is available from Acrison, Inc., as model MDII® microprocessor-based weigh feeder controller.
The controller 12 is operatively engaged to the refill valve 21 via valve control line 210. The controller 12 can
produce a valve control signal to open or close the refill valve 21 . When the refill valve 21 is open, material can flow
40 from the refill hopper 20 through the outlet aperture 23 to the supply vessel 2, thereby refilling the supply vessel 2.
The high level probe 30 is operatively engaged to the controller 12 via probe data line 300. The high level probe
30 produces a "high level" electrical signal to the controller 12 which indicates whether material is located in the supply
vessel 2 at the level L of the probe 30.
The controller 12 is operatively engaged to the scale 11 via scale data line 110 such that the scale 11 can send a
45 weight signal to the controller 12 representative of at least the weight of the material in the supply vessel 2.
The controller 12 is operatively engaged to the feeding assembly 10 via assembly control line 100. The controller
12 is capable of producing a motor command signal to the motor 13 of the feeding assembly 10 to control the feed
rate by either increasing, decreasing or maintaining the motor speed. A shaft encoder 16 is connected to the motor 13
and generates an actual motor speed signal to the controller 12 via a motor speed line 101, indicative of the actual
so speed of the motor 13. In this way, the controller 12 keeps track of the present speed of the motor. Further, every
rotation of the auger 15 displaces a known volume of material. The actual motor speed signal therefore corresponds
to a displaced volume of material and the shaft encoder 16 also operates as a volumetric displacement meter.
The loss-in-weight feeding system is designed such that the weight of the entire loss-in-weight feeding system,
namely, the feeding assembly 10, and its supply vessel 2, are tared-off or counter-balanced so that the scale 11 weighs
55 only the material, producing an electrical weight signal to the controller 12 via scale data line 110 indicative of the
weight of material measured. The controller 12 compares the weight measured at different times to determine the "loss-
in-weight," i.e., the weight of material fed by the feeding system over a period of time. In this way, the actual weight of
the material fed is measured. This loss-in-weight is compared to an expected or "set" loss-in-weight, causing the gen-

4
 EP0 631 115 B1

eration of the motor command signal by controller 12. The command signal is transmitted to the motor 13 via assembly
control line 100 to affect the motor speed, either increasing, maintaining, or decreasing it, to achieve the set weight
loss. Control of a loss-in-weight feeding system in such a manner is well-known.
As material is fed from the feeding assembly 10, the total weight decreases. When the weight measured by the
5 scale 11 reaches a predetermined low level, the controller 12 produces a valve control signal along valve control line
21 0 to actuate the refill valve 21 . This opens refill valve 21 and allows material to flow into the supply vessel 2. Alter-
natively, a low level probe 40 can be placed at a low level in the supply vessel 2. The low level probe 40 produces a
signal to the controller 12 via low level probe line 400 indicating that there is no material in the supply vessel 2 at the
low level. The controller 12 then produces the valve control signal to open valve 21 and begin refill. In either case, the
10 weight measured by scale 11 consequently increases.
In existing loss-in-weight feeding systems, the weight signal must reach a predetermined high level signal (from
the scale) before the controller 12 sends a signal to close the refill valve and cease refill. During refill, the system no
longer operates in a loss-in-weight mode, but rather operates in a volumetric mode.
Existing feeding systems can have problems arising from the varying density of the refilled material which may
is range in density from 30 to 400% of the density of the material 3 originally in the supply vessel 2. Such density variations
cannot be easily predicted or controlled by an operator and occur even though the refilled material has the same
composition. When the refill material has a very low bulk density, the predetermined high level weight signal may never
be reached. Consequently, the refill valve 21 will remain open such that material overfills the supply vessel 2 during
refill. Further, during refill, the loss-in-weight feeding system is operating in a volumetric mode which is generally less
20 accurate than the gravimetric or loss-in-weight mode. Consequently, increasing the time spent in refill decreases the
accuracy of the feeding system.
The varying density of the material can cause other problems in the operation of the feeding system. The controller
12 produces a signal to the control panel 31 indicative of the volume of the supply vessel filled with material based, in
part, on the density of the material. This volume is displayed on the panel display 32. When the material density varies
25 without a corresponding correction in the controller 12, the displayed material volume is less accurate.
When the controller 12 must invoke the "fast start" routine as described above in the Background of the Invention,
the estimated motor speed for the new feed rate is determined based on the maximum feed rate which, in turn, is
based in part on the bulk density of the material. When this density changes, the controller's estimate of the new motor
speed is less accurate.
30 When the feeding system is operating in a volumetric mode, the feed rate or material output is determined based,
in part, on the bulk density of the material. When the bulk density of the material varies without a corresponding cor-
rection in the controller 12, the determination of material fed in volumetric mode is less accurate. Consequently, updating
the bulk density of the material in the controller 12 provides many benefits to the operation of a weight-loss feeding
system for producing optimum performance.
35
Automatic Refill Shut Off

FIG. 2 is a block diagram showing a portion of the method of controller 12 during refill of the supply vessel 2. In
the system of the present invention, refill begins when the weight signal from scale 11 reaches a predetermined low
40 level settings (or when low level probe 40 indicates that there is no material at a low level in the supply vessel 2).
Controller 12 then signals refill valve 21 to open, beginning refill. Material flows from the refill supply source, such as
refill hopper 20 into the supply vessel 2. Refill is continued until the weight signal from the scale 11 reaches a prede-
termined high level setting or when high level probe 30 senses material 3 in the supply vessel 2 at the predetermined
high level L (box 50). If the predetermined high level weight setting in supply vessel 2 is not reached due to a lower
45 density product, the high level probe 30 produces a high level signal to the controller 12 along the probe data line 300,
indicating that the supply vessel 2 is filled to level L. The controller 12 then sends a signal to the refill valve 21 along
valve control line 210, closing the valve 21 and stopping refill (box 51). Consequently, material does not overfill the
supply vessel 2 and the feeding system can return to loss-in-weight mode. The controller 12 is directed through this
operation by a section of the control program which is reproduced below as the Automatic Refill Shut Off Routine.
50
Present Bulk Density Calculating

In accord with another aspect of the present invention, the bulk density of the material is recalculated after refill
based on the actual weight and volume of the refill material in the supply vessel 2. The present bulk density is preferably
55 calculated directly after refill so that the high level probe 30 can operate as a refill shut off mechanism as well as a
material volume indicator. If the high level probe 30 was disposed lower in the supply vessel 2, the bulk density could
still be calculated in accord with the present invention so long as the weight of the material in the supply vessel 2 is
measured at the time that the probe 30 detects material corresponding to a predetermined volume. Since neither

5
 EP0 631 115 B1

volumetric feeding systems nor weigh-belt/weigh-auger feeding systems generally come equipped with a scale capable
of determining the weight of the material in its supply vessel 2, this aspect is intended presently for use with loss-in-
weight feeding systems.
The material level L sensed by the high level probe 30 corresponds to a predetermined volume in the supply vessel
5 2 up to that level, i.e., "PROBE VOLUME" in FIG. 2, box 52. When the high level probe 30 detects material at level L,
the volume of material in the supply vessel 2 is equal to the probe volume (or a full supply vessel 2 of a known volume).
The scale 11 produces an electrical weight signal corresponding to the weight of the material in the entire the feeding
system. The weight signal is transmitted to the controller 12 via scale data line 110. The weighing mechanism is de-
signed such that the weight of the feeding system is eliminated, thus producing a weight signal corresponding solely
10 to the net weight of the material in the feeding system. Knowing the type and size of the supply vessel 2 and the
metering mechanism 14, the portion of the total weight attributable to the material in the supply vessel, i.e., "SCALE
WEIGHT" in FIG. 2, box 52, can be determined in a known manner. If the bulk density has been calculated, the controller
12 returns to the main loop (box 59). If not, the controller 12 then calculates the present bulk density of the refilled
material according to the following equation (box 52):
15
PRESENT BULK DENSITY- ^/g^ (1)

This calculation is performed by the section of the control program reproduced below as the Bulk Density Calculation
20 Routine. After this calculation, the controller 12 returns to the main control loop.

Controller & Display

The controller 12 commands the panel display 32 to indicate the volume of the supply vessel 2 filled with material.
25 The controller 12 determines the weight of the material in the supply vessel 2 as explained above. Using the present
bulk density of the material, the controller 12 can determine the total volume of material in the supply vessel 2 in accord
with the following equation:

SCALE WEIGHT
MATERIAL VOLUME = (2)
PRESENT BULK DENSITY

The material volume is then divided by the total volume of the supply vessel 2, resulting in the percentage of supply
vessel volume filled with material. The controller 12 commands the panel display 32 to display this percentage. This
can be displayed continuously, at set time intervals, at the request of the operator or in other manners.
35
Volumetric Mode

Once the present bulk density of the material is calculated, it is used to affect the control of the feeding assembly
10 and the calculation of the amount of material fed. During operation of the loss-in-weight feeding system 1, the
40 controller 12 will cause the feeding assembly 10 to operate in a volumetric mode at various times. This will occur,
among other times, during refill, and also during a disturbance, such as an external sharp impact against the feeding
system or collapsing bridged material in the supply vessel of a dry solids feeding system when the system enters, for
example, the "Acri-Lok" mode, as explained, for example, in U.S. Patents Re. 30,967, Re. 32,101 and Re. 32,102.
At all times, the controller 12 keeps track of the material feed rate to maintain the operator-input set rate. In the
45 volumetric mode, the controller 12 determines the feed rate based on the volume of the material fed and the bulk
density of the material being fed. In a preferred system, the feeding assembly 10 includes an auger metering mechanism
14 with a feed auger 15 driven by a motor 13. A shaft encoder 16 is operatively engaged to the motor and produces
a signal corresponding to the motor speed, and thus the shaft speed. The auger 15 displaces a known volume of
material per revolution of the motor ("auger displacement"). The shaft encoder 16 generates a signal corresponding
50 to the actual motor speed and transmits the signal to the controller 12 via motor speed line 101 . Consequently, controller
12 can determine the feed rate of material displaced by the auger 15 during a given period of operation from the actual
motor speed signal received from the shaft encoder 16 in accord with the following equation:

55 ' FEED RATE I N

AUGER ACTUAL PRESENT
VOLUMETRIC \ MOTOR BULK (3)
MODE DISPLACEMENT J
SPEED DENSITY

6
 EP0 631 115 B1

As the feed rate of material displaced by the auger 15 (i.e., weight of material fed over a period of time) becomes
known, it is compared to the set rate. The controller 12, in turn, generates a motor command signal to increase, maintain,
or decrease motor speed to match material discharge to the set rate. Changing the bulk density to a more accurate
measurement increases the accuracy of the feeding system when it shifts to a volumetric mode for whatever reason
5 (refill, disturbance or fast start) and, thus, enhances the overall accuracy of the feeding system.

Fast Start

At times during operation of a loss-in-weight feeding system, standard loss-in-weight control will not get "on rate"
10 as quickly as is desired. That is, standard loss-in-weight control will not change the motor speed to reach a desired
set rate fast enough to achieve the required feed rate and/or metering accuracy, particularly when first turned on. To
address this problem, the controller 12 is equipped with a "fast start" routine that quickly adjusts the motor speed to
quickly achieve the new set rate. Optionally, the controller can be designed such that, when the operator inputs a new
desired set rate which differs from the present set rate by a predetermined amount (e.g., 20%), the fast start routine
is will be actuated. The predetermined amount should be adjustable by the operator.
Assuming that feed rate and motor speed are linearly related from no or zero feed rate at zero motor speed up to
a maximum feed rate at 100% of motor speed, the controller 12 determines the speed at which the motor 13 should
run based on the relationship of the desired set rate to maximum feed rate. If the maximum feed rate (i.e., the feed
rate if the motor was operating at 100% of speed) is 100 Ibs/hr and the new set rate is 75 Ibs/hr, the controller 12 will
20 signal the motor 13 via assembly control line 100 to run at 75% of maximum speed. Once the motor 13 has been
operating at the new speed for a sufficient time, loss-in-weight control can continue.
The maximum feed rate, however, varies as the bulk density of the material changes. In existing feeding systems,
the present bulk density of the material being fed is estimated, for example, every 15 seconds based on the present
motor speed and the present feed rate as explained below.
25 In the system of the present invention, bulk density can be calculated directly and used to adjust the maximum
feed rate as is necessary. As shown in FIG. 3, which is a block diagram of the maximum rate/bulk density auto com-
pensation function, the controller 12 determines whether there is material on the high level probe 30 (box 53). If so, a
present bulk density is calculated in accord with eq. 1, above (box 54). The present maximum feed rate is then deter-
mined in accord with the following equation (box 55):
30

PRESENT MFR = ( PRESENT BULK DENSITY^ ^ ^ ^ " ^ J K«)

\ PRIOR BULK DENSITY j

35
The prior bulk density and the prior maximum feed rate were stored by the controller 12 from the previous determination
of these values. This new maximum feed rate is used in the "fast start" calculation.
If material is not detected by the high level probe 30, the controller 12 continues with standard operation. When
the controller 12 determines that the timer is at 15 seconds (box 56), the bulk density is calculated by a portion of the
40 control program reproduced below as the Max Rate/Bulk Density Auto Compensation Routine. Initially, it is assumed
the feeding system is feeding material at the control set point input by the operator. Then, the controller 12 determines
the speed at which the motor 13 is presently running (from shaft encoder 16) as a percentage of the maximum motor
speed. This percentage is the present motor speed. A present maximum feed rate (or present mfr) is calculated in
accord with the following equation (box 57):
45

PRESENT MFR = / CONTROL SET POINT \

\ PRESENT MOTOR SPEED j (5)
50

Optionally, limits can be placed on the present maximum feed rate such that no drastic changes occur in the control
of the feeding system. One acceptable limit on the range of the present maximum feed rate is within 2.5% of the prior
maximum feed rate. If the present maximum feed rate as calculated in accord with eq. 5 is outside that range, the
55 present maximum feed rate can be set equal to the upper or lower limit, depending on whether the calculated present
feed rate is above or below the range, respectively. Also, the calculation of present maximum feed rate is not performed
if a feed rate deviation alarm is present, a refill is in progress or has just completed, or an Acri-Lok alarm is present.
Once the present maximum feed rate is determined, the present bulk density is calculated in accord with the

7
 EP0 631 115 B1

following equation (box 58):

PRESENT BULK PRESENT MFR PRIOR BULK

DENSITY PRIOR MFR DENSITY (6)

The present bulk density is used, as explained above, in the control of the feeding system during operation in a volu
metric mode and to control the calculation of the volume of material in the supply vessel 2.
As can be seen, applicants' invention permits more accurate display of material volume in the supply vessel 2
more accurate operation of the feeding system, and more accurate estimates of the new motor speed in fast start.

Automatic Refill Shut Off Routine

One example of an automatic refill shut off routine written in C language is shown below.

Copyright o 1992 Acrison, Inc.

Al I Rights Reserved
void Ref i ll(void)
{
static int Ref i I l_Delay_Timer, Refill_Turnoff_Timeout;
static int Dev i a t i on_Ref i U_B Iock , Pul sed_T i mer;
if (Pulsed_Timer)
{
System.Pulsed_Output = TRUE;
if (--Pulsed_Timer == 0)
System. PulsedjDutput = FALSE;

8
 EP0 631 115 B1

t
if we ire fn Batch In, or have No Scale, or are in External Fast node,
are in Weigh Vessel node, or High Level Probe, we cancel (force off) refills.
if (System.Batch.In || System.Mo_Scale_Signal || System.HodeS !! Batch Direction Setpoint ■» 2
System. High Level Probe) " "
(
Turn_Of f_Ref i I KiDeviat ion_Ref i I l_Block);
~
return;
>
if (Refill_Delay_Timer) // if MReful De[ay« j, set...
if (--Refill_Delay_Timer «« 0) // And reaches lero...
C
Previous_Totalfed » Total_Fed;
Previous~Totalfed_2 « Total Fed 2;
Previous_Batch_Fed « Batch_7ed;~
Previous_Scale_Ueight * Scale Weight In Units;
Previous_Scale_Weight_2 » ScaTe_Weigh"t_Tn_Units;
Previous_Scale_Weight_Batching * Scale~WeTght ~In~Units;
Start_Band * SECONDS(Start_Band_Setpoint);
~ // We go into Fast Start
Aerilok_Timer * 0; // and set all other flags
Alarm.8it.Acrilok = FALSE; // accordingly
Alarm. Bit. Deviation = FALSE;
if (Compensation Selection)
<
High_Level_Sample_Delay > SECONDSC60);
Compensation_Average_Period ■ SECONDS(5);
Compensation_Level_Svjn • 0;
Compensation_FeedRate_Sum = 0;
Compensation_HotorSpeed_Sum ■ 0;

it external input refill contacts close. Go into refill. But set a flag
so you only do it once. Reset flag when contact reopens.
if (System. External Refill Contact)
C
if (ISystem. Digital In_Refill Request)
{
Ref i I l_0N< );
System. Digital_In_Ref ill Request = TRUE;
>
else
System. Digital In Refill Request * FALSE;
►
If power was just turned on a few seconds ago, or refill is being
:anceled, shut it off.
if (ignore_Readings |j System. Hanual_Ref i I t_Cancel )
Turn_Of f_Ref i 11(tDeviat i on_Ref
~ i I l_Block)
~ ;
return;
►
f the following self explanatory condition is true, a refill request will be latched.
if (S1.Bit.low_Level IS. S1.Bit.Auto_Fill It, !System. Low_Level_Latched U
System. Prevent_Ref i I l_Restart) System.Low_Level_Latched * TRUE;
f (S1.Bit.Refill_ON) // if in refill...
Ref ill_Delay_Timer = SECOHDS(Fast_Af ter Refill); // Set the Refill Delay timer
8atch_Settle_Delay = SECOMDS(Batch_Start Delay); // Set the Batch Settling delay timer
if (Refill_Turnoff_Timeout) // if the Refill timeout counter is set.,
if (--Refill_Turnoff_Tirneout « 0) // and it reaches zero...
 EP0 631 115 B1

TurnOf f_Ref i U (&Devi at i on_Ref i I l_B t oek) ; // Shut off Refill

Error_Number « 8; // Throw up the error message
System. Ref ill Expired >= TRUE; // Set status flag
S3.Bit.Refill~Expired = TRUE;
PulsedJTimer ■ SECONDSC10); // Start "pulsed" output contact
if (Stop_On_Empty_Or_RTO) // and if cal entry says to STOP,
Emergency_Stop(); // we STOP
return;

// When we reach High Level OR Overfill we shut off Refill.

if (S1. Bit. High Level || SZ.Bit.Overf i 11)
C
Turn_Off_Ref i I l(SOeviation_Ref i I ^Blocks-
return;
>
// If a manual refill is requested, we set the request here.
if (System. Hanualjtef i ll_Request)
{
System. Low_Level_Latched = TRUE;
System. Manual_Ref ill_Request = FALSE;
>
// If no refill requests are pending, we are all done in this function.
if (! System. Low_Level_Latched)
return;
// If we batching and we dont have Batch complete and are running and
// Batch through Refill is off, we Stop, set the restart timer, and wait for
// the Totalizer delay to expire. Then we go into refill.
if (Batch Flags. Batching)
if (!Batch_Flags.Batch_Complete && System. Running && Batch_Thru_Ref i U == 0)
<
Emergency_Stop( ) ;
Batch_Waiting_To Restart = SECONDS(25);
}
if (Batch_Total_Delay && Batch_Thru_Ref i 11 == 0)
return;

// If we have a Deviation at the time Refill is requested, we delay going

// into refill in the hopes the deviation will end. If it doesn't end
// before the timer expires, ue go into refill anyway.
if (S2. Bit. Deviation)
if (Deviation Ref i I l_Block)
(
--Deviation_Ref i I l_Block;
return;
>
// If we pass all conditions and get to this point, ue start a Refill.
Turn_On_Ref i 11(&Ref i I l_Turnof f_T imeout );

Bulk Density Calculation Routine

One example of a bulk density calculation routine written in C language is shown below.

10
 EP0 631 115 B1

Copyright e 1992 Acrison, Inc.

All Rights Reserved
void High Level Probe(void)
{
static int done;
if <Systeni.0i9ital_Inputs_Used It System. High_Level_Probe It Probe_Volume U done " 0)
t
Bulk_Density ■ Scale_Weight_In_Units / Probe_volume;
done « -1;
>
if (I System. High_Level_Probe)
done * 0);
>

Max Rate/Bulk Density Auto Compensation Routine

One example of a max rate/bulk density auto compensation routine written in C language is shown below.

Copyright « 1992 Acrison, Inc.

All Rights Reserved
void Compute Change_To_Maxrate(void)
(
static int timer = SEC0NDSO5);
double Motor_Correction, upper_limit, lower_limit, temp;
if (ISystem.Running |j S1.Bit.Refi I l_ON J| Alarm.Current_Alarms jj S2.Bit.Fast_Start J|
System. High_Level_Probe || Control_Mode > timer = SECON0SC15);
if (Compensation_Se lection ==4)
return;
if (--timer)
return;
timer « SECONDS(15);
temp * Max_Feed_Rate * .025;
upper_limit « temp ♦ Hax_Feed_Rate;
lower~limit = Max_Feed_Rate - temp;
temp * Control_Setpoint / (Motor_Command / 100);
if (temp > upper_limit)
Max_Feed_Rate * upper_limit;
else if~(temp < lower limit)
Max_Feed_Rate = lower_limit;
else
Hax_Feed_Rate * temp;
Mis_Feed_Rate « Max_Feed_Rate * .5;

if (Compensation_Se lection I • 3)
Bulk_Oensity » (Max_Feed_Rate / Prior_Max_Feed_Rate) " Prior_Bulk_Density
Prior_Hax_Feed_Rate = Max_Feed_Rate;
Prior_Mid_Feed_Rate = Mid_Feed_Rate;
Prior~Bulk_Density = Bulk~Dens7ty;
Last_Bulk_Density = Bulk_Density;

Claims

1. A method of operating a loss-in-weight feeding system for feeding material (3) at a controllable feed rate, compi
the steps of:

11
 EP0 631 115 B1

1. 1 filling a supply vessel (2) with material (3) to be fed;

1.2 weighing at least the material (3) in the supply vessel (2);
1.3 sensing the level of the material (3) in the supply vessel (2);
1.4 calculating the present bulk density of the material (3) based on the level of the material (3) and the weight
5 of the material (3);
1.5 feeding material (3) from the supply vessel (2) at a controllable feed rate based at least in part on the
present bulk density of the material (3).

2. The method of claim 1 wherein the step of feeding material (3) comprises rotating an auger (1 5) at a rotation rate
10 within a metering system at a controllable rotation rate wherein the auger (15) is capable of displacing a volume
of material (3).

3. The method of claim 2 further comprising the step of calculating the feed rate based, at least in part, on the rotation
rate.
15
4. The method of claim 3 further comprising the step of determining the weight of the material (3) fed based on the
volume of material (3) displaced and the bulk density of the material (3).

5. The method of claim 1 further comprising the step of calculating the volume of material (3) in the supply vessel (2)
20 based on the weight of the material (3) and the present bulk density of the material (3).

6. The method of claim 5 further comprising the step of displaying the volume of material (3) in the supply vessel (2).

7. The method of claim 1 further comprising refilling the supply vessel (2) when a pre-determined amount of material
25 (3) remains in the supply vessel (2).

8. The method of claim 7 further comprising the step of stopping refill when a pre-determined level of material (3) is
sensed.

30 9. The method of claim 1 further comprising the steps of

9.1 generating a high level signal corresponding to the level of material (3);
9.2 transmitting the high level signal to a controller (12);
9.3 weighing the feeding system and generating a weight signal corresponding at least to the weight of material
35 (3) in the supply vessel (2);
9.4 transmitting the weight signal to a controller (1 2);
9.5 calculating a maximum feed rate based in part on the high level signal and the weight signal; and
9.6 controlling the feed rate based in part on the maximum feed rate by generating and transmitting a control
signal from the controller (1 2) to the motor (1 3) so that the motor (1 3) runs at a desired speed.
40
10. The method of claim 9 wherein the desired speed is determined by multiplying the maximum speed of the motor
(1 3) by the ratio of the selected feed rate and the maximum feed rate.

11. The method of claim 9 further comprising the step of refilling the supply vessel (2) when a pre-determined amount
45 of material (3) remains in the supply vessel (2).

12. The method of claim 1 further comprising the steps of

12. 1 placing material (3) to be fed having a bulk density into the supply vessel (2);
so 12.2 releasing material (3) from the supply vessel (2) into a feeding mechanism (10);
12.3 displacing material (3) from the feeding mechanism (10) at a controllable displacement rate;
12.4 refilling the supply vessel (2) with new material when a pre-determined amount of material (3) remains
in the supply vessel (2);
12.5 weighing at least the material (3) in the supply vessel (2) wherein the steps of weighing and level sensing
55 are performed on the same material (3); and
12.6 calculating the bulk density of the new material (3) in the supply vessel (2) based at least in part on the
weight of the new material (3) in the supply vessel (2) and the level of the new material (3) in the supply vessel
(2).

12
 EP0 631 115 B1

13. The method of claim 12 further comprising the steps of stopping refill when the material (3) reaches a pre-deter-
mined level.

14. Device for carrying out the method of one or more of the preceding claims, comprising:
5
a) a material supply vessel (2)
b) a discharge mechanism or feeding assembly (10) at the bottom of said supply vessel (2)
c) a refill supply source (20) positioned above the supply vessel (2)
d) a refill valve (21 ) disposed at a bottom outlet aperture (23) of said refill supply source (20)
10 e) a weighing mechanism or scale (11 ) disposed adjacent to said feeding assembly (10)
f) a high level probe (30)
g) a low level probe (40)
h) a controller (12) operatively engaged to the high level probe (30), to the low level probe (40), to the weighing
mechanism or scale (11) and to the feeding assembly (10), the controller (12) being equipped in such a manner
is that it calculates the present bulk density of the material (3) based on the level of the material (3) and the
weight of the material (3).

Patentanspriiche
20
1. Verfahren zum Betatigen eines Abmangel-Zuf uhrsystems zum Zuf uhren von Material (3) in einer steuerbaren Zu-
fuhrmenge, das folgende Schritte umfaBt:

1.1 Fullen eines Vorratsbehalters (2) mit zuzufuhrendem Material (3);

25 1.2 Wiegen zumindest des in dem Vorratsbehalter (2) befindlichen Materials (3);
1.3 Erkennen des Fullstands des Materials (3) in dem Vorratsbehalter (2);
1.4 Errechnen der aktuellen Schuttdichte des Materials (3) basierend auf dem Fullstand des Materials (3) und
dem Gewicht des Materials (3);
1.5 Zufuhren von Material (3) aus dem Vorratsbehalter (2) in einer steuerbaren Zufuhrmenge zumindest teil-
30 weise basierend auf der aktuellen Schuttdichte des Materials (3).

2. Verfahren nach Anspruch 1, bei dem der Schritt des Zufuhrens von Material (3) das Drehen eines Bohrers (15)
mit einer Umdrehungszahl innerhalb eines Mel3-systems bei einer steuerbaren Umdrehungszahl enthalt, wobei
der Bohrer (15) ein Materialvolumen (3) bewegen kann.
35
3. Verfahren nach Anspruch 2, weiterhin umfassend den Schritt des Errechnens der Zufuhrmenge auf der Basis
zumindest teilweise der Umdrehungszahl.

4. Verfahren nach Anspruch 3, weiterhin umfassend den Schritt des Feststellens des Gewichts des zugefuhrten Ma-
40 terials (3), basierend auf dem geforderten Materialvolumen (3) und der Schuttdichte des Materials (3).

5. Verfahren nach Anspruch 1, weiterhin umfassend den Schritt des Errechnens des Volumens des Materials (3) in
dem Vorratsbehalter (2), basierend auf dem Gewicht des Materials (3) und der aktuellen Schuttdichte des Materials
(3).
45
6. Verfahren nach Anspruch 5, weiterhin umfassend den Schritt des Anzeigens des Volumens des Materials (3) in
dem Vorratsbehalter (2).

7. Verfahren nach Anspruch 1, weiterhin umfassend das Wiederbefullen des Vorratsbehalters (2), wenn ein vorbe-
50 stimmter Betrag des Materials (3) in dem Vorratsbehalter (2) verbleibt.

8. Verfahren nach Anspruch 7, umfassend den Schritt des Beendens des Befullens, wenn ein vorbestimmter Fullstand
des Materials (3) ermittelt wird.

55 9. Verfahren nach Anspruch 1, weiterhin umfassend die Schritte des

9.1 Erzeugens eines Hoher-Fullstand-Signals, das dem Fullstand des Materials (3) entspricht;
9.2 Ubermittelns des Hoher-Fullstand-Signals zu einem Regler (12);

13
 EP0 631 115 B1

9.3 Wiegens des Zufuhrsystems und Erzeugens eines Gewichtsignals, das zumindest dem Gewicht des Ma-
terials (3) in dem Vorratsbehalter (2) entspricht;
9.4 Ubermittelns des Gewichtsignals zu einem Regler (12);
9.5 Errechnens einer maximalen Zufuhrmenge teilweise basierend auf dem Hoher-Fullstand-Signal und dem
5 Gewichtsignal; und
9.6 Steuerns der Zufuhrmenge teilweise basierend auf der maximalen Zufuhrmenge durch Erzeugen und
Ubermitteln eines Steuersignals durch den Regler (12) zu dem Motor (13), so dal3 der Motor (13) mit einer
gewunschten Geschwindigkeit lauft.

10 10. Verfahren nach Anspruch 9, bei dem die gewunschte Geschwindigkeit bestimmt wird durch Multiplizieren der ma-
ximalen Geschwindigkeit des Motors (1 3) mit dem Verhaltnis aus ausgewahlter Zufuhrmenge und maximaler Zu-
fuhrmenge.

11. Verfahren nach Anspruch 9, weiterhin umfassend den Schritt des Wiederbefullens des Vorratsbehalters (2), wenn
is ein vorbestimmter Betrag des Materials (3) in dem Vorratsbehalter (2) verbleibt.

12. Verfahren nach Anspruch 1, weiterhin umfassend die Schritte des

12.1 Eingebens von zuzufuhrendem Material (3) mit einer Schuttdichte in den Vorratsbehalter (2);
20 12.2 Ablassens des Materials (3) von dem Vorratsbehalter (2) in einen Zufuhrmechanismus (10);
12.3 Fortbewegens des Materials (3) von dem Zufuhrmechanismus (10) mit einer steuerbaren geforderten
Menge;
12.4 Wiederbefullens des Vorratsbehalters (2) mit neuem Material, wenn eine vorbestimmte Materialmenge
(3) in dem Vorratsbehalter (2) verbleibt;
25 12.5 Wiegens zumindest des Materials (3) in dem Vorratsbehalter (2), wobei die Schritte des Wiegens und
Fullstand-Fuhlens an demselben Material (3) vorgenommen werden; und
12.6 Errechnens der Schuttdichte des neuen Materials (3) in dem Vorratsbehalter (2), wenigstens teilweise
basierend auf dem Gewicht des neuen Materials (3) im Vorratsbehalter (2) und dem Fullstand des neuen
Materials (3) in dem Vorratsbehalter (2).
30
13. Verfahren nach Anspruch 12, ferner umfassend die Schritte des Beendens des Wiederbefullens, wenn das Material
(3) einen vorbestimmten Fullstand erreicht hat.

14. Vorrichtung zum Durchfuhren des Verfahrens nach einem oder mehreren der vorhergehenden Anspruche, um-
35 fassend

a) einen Vorratsbehalter (2)

b) einen Entlademachanismus oder eine Zufuhreinrichtung (10) am Boden des Vorratsbehalters (2)
c) eine Wiederbefullungs-Vorratsquelle (20), die oberhalb des Vorratsbehalters (2) angebracht ist
40 d) ein Wiederbefullungs-Ventil (21), das am Boden der AuslaBoffnung (23) der Wiederbefullungs-Quelle (20)
angebracht ist
e) einen Mechanismus zum Wiegen oder einen MaBstab (11), der an die Zufuhreinrichtung (10) angrenzend
angeordnet ist
f) einen Hoher-Fullstand-Fuhler (30)
45 g) einen Niedriger-Fullstand-Fuhler (40)
h) einen Regler (1 2), der mit dem Hoher-Fullstand-Fuhler (30), dem Niedriger-Fullstand-Fuhler (40), dem Me-
chanismus zum Wiegen oder MaBstab (10) und zu der Zufuhreinrichtung (10) in operativem Eingriff steht,
wobei der Regler (12) so ausgestattet ist, dal3 er die aktuelle Schuttdichte des Materials (3) basierend auf
dem Fullstand des Materials (3) und dem Gewicht des Materials (3) errechnet.
50

Revendications

1. Un procede de fonctionnement d'un systeme d'alimentation differentiel pour alimenter une matiere (3) a un debit
55 pouvant etre commande, comprenant les operations consistant a :

1.1. remplir un reservoir (2) avec une matiere (3) a alimenter ;

1.2 peser au moins la matiere (3) se trouvant dans le reservoir (2);

14
 EP0 631 115 B1

1.3 detecter le niveau de la matiere (3) dans le reservoir (2) ;

1.4 calculer la masse volumique apparente actuelle de la matiere (3) sur la base du niveau de la matiere (3)
et du poids de la matiere (3);
1.5 alimenter la matiere (3) depuis le reservoir (2) a un debit pouvant etre commande sur la base, au moins
en partie, de la masse volumique apparente actuelle de la matiere (3).

2. Le procede de la revendication 1 dans lequel I'operation consistant a alimenter la matiere (3) comprend la mise
en rotation d'une meche (15) a une vitesse de rotation a I'interieur d'un systeme de mesure a une vitesse de
rotation pouvant etre commandee, ou la meche (15) est capable de deplacer un volume de matiere (3).

3. Le procede de la revendication 2 comprenant en outre I'operation consistant a calculer la vitesse d'alimentation

sur la base, au moins en partie, de la vitesse de rotation.

4. Le procede de la revendication 3 comprenant en outre I'operation consistant a determiner le poids de la matiere

(3) amenee sur la base du volume de matiere (3) deplacee et de la masse volumique apparente de la matiere (3).

5. Le procede de la revendication 1 comprenant en outre I'operation consistant a calculer le volume de matiere (3)
dans le reservoir (2) sur la base du poids de la matiere (3) et de la masse volumique apparente actuelle de la
matiere (3).

6. Le procede de la revendication 5 comprenant en outre I'operation consistant a afficher le volume de matiere (3)
dans le reservoir (2).

7. Le procede de la revendication 1 comprenant en outre le fait de recharger le reservoir (2) lorsqu'une quantite
predeterminee de matiere (3) reste dans le reservoir (2).

8. Le procede de la revendication 7 comprenant en outre I'operation consistant a arreter le rechargement lorsqu'un

niveau predetermine de matiere (3) est detecte.

9. Le procede de la revendication 1 comprenant en outre les operations consistant a

9.1. produire un signal de niveau haut correspondant au niveau de matiere (3) ;

9.2. transmettre le signal de niveau haut a un organe de commande (12) ;
9.3. peser le systeme d'alimentation et produire un signal de poids correspondant au moins au poids de matiere
(3) dans le reservoir (2);
9.4 transmettre le signal de poids a un organe de commande (12) ;
9.5 calculer une vitesse d'alimentation maximum sur la base, en partie, du signal de niveau haut et du signal
de poids ; et
9.6 commander la vitesse d'alimentation sur la base, en partie, de la vitesse d'alimentation maximum en pro-
duisant et en emettant un signal de commande a partir de I'organe de commande (12) vers le servomoteur
(1 3) de sorte que le moteur (1 3) tourne a une vitesse desiree.

10. Le procede de la revendication 9 dans lequel la vitesse desiree est determinee en multipliant la vitesse maximum
du moteur (13) par le rapport de la vitesse d'alimentation choisie et de la vitesse d'alimentation maximum.

11. Le procede de la revendication 9 comprenant en outre I'operation consistant a recharger le reservoir (2) lorsqu'une
quantite predeterminee de matiere (3) reste dans le reservoir (2).

12. Le procede de la revendication 1 comprenant en outre les operations consistant a

12.1 placer la matiere (3) a alimenter, qui presente une masse volumique apparente, dans le reservoir (2) ;
12.2 decharger la matiere (3) du reservoir (2) dans un mecanisme d'alimentation (10) ;
12.3 deplacer la matiere (3) depuis le mecanisme d'alimentation (10) a une vitesse de deplacement pouvant
etre commandee ;
12.4 recharger le reservoir (2) en nouvelle matiere lorsqu'une quantite predeterminee de matiere (3) reste
dans le reservoir (2) ;
12.5 peser au moins la matiere (3) se trouvant dans le reservoir (2), ou les operations de pesee et de detection
de niveau sont effectuees sur la meme matiere (3) ; et

15
 EP0 631 115 B1

12.6 calculer la masse volumique apparente de la nouvelle matiere (3) se trouvant dans le reservoir (2) sur
la base, au moins en partie, du poids de la nouvelle matiere (3) se trouvant dans le reservoir (2) et du niveau
de la nouvelle matiere (3) se trouvant dans le reservoir (2).

13. Le procede de la revendication 12 comprenant en outre les operations consistant a arreter le rechargement lorsque
la matiere (3) atteint un niveau predetermine.

14. Dispositif pour mettre en oeuvre le procede d'une ou plusieurs des revendications precedentes, comprenant :

a) un reservoir de matiere (2)

b) un mecanisme de decharge ou un ensemble d'alimentation (10) au fond dudit reservoir (2)
c) une source de rechargement (20) placee au dessus du reservoir (2)
d) une vanne de rechargement (21) disposee au niveau d'une ouverture de sortie inferieure (23) de ladite
source de rechargement (20)
e) un mecanisme de pesee ou une balance (11) disposee de maniere adjacente audit ensemble d'alimentation
(10)
f) un capteur de niveau haut (30)
g) un capteur de niveau bas (40)
h) un organe de commande (12) engage de maniere operationnelle avec le capteur de niveau haut (30), avec
le capteur de niveau bas (40), avec le mecanisme de pesee ou balance (11) et avec I'ensemble d'alimentation
(10), I'organe de commande (12) etant equipe de telle maniere qu'il calcule la masse volumique apparente
actuelle de la matiere (3) sur la base du niveau de la matiere (3) et du poids de la matiere (3).

16
EP0 631 115 B1

F I G . I

17
 EP0 631 115 B1

FROM MAIN
CONTROL L O O P

ENTER REFILL
SUBROUTINE

51

STOP REFILL
(SIGNAL TO REFILL
VALVE ALONG 2 1 0 )

CONTINUE
/ HAS X
WITH REFILL
uONTROLLErO
YES COMPUTED
NEW BULK
v DENSITY /
\ YET?/

52

P R E S E N T BULK DENSITY =
SCALE W E I G H T / P R O B E V O L U M E

CONTINUE WITH
MAIN CONTROL L O O P

FIG. 2

18
I