Patent Grant
8768632
The present invention relates to a method and an apparatus serving to fill target containers with a predetermined target quantity of a free-flowing substance out of a reservoir.
Filling devices of this kind are used in particular in the dispensing of small dosage amounts as required for example in the field of pharmaceuticals. The target containers are often set on a balance in order to weigh the amount of substance delivered by the dosage-dispensing device, so that the substance can subsequently be further processed in accordance with given instructions.
The substance to be dispensed is held for example in a source container or reservoir which is equipped with a dispensing head. It is desirable if the substance to be dispensed is delivered to the outside through an opening of the dosage-dispensing device, so that at the end of the filling process a predetermined target mass of substance will have been received by the target container. It is of importance here that the actual amount of mass in the target container agrees as accurately as possible with the predetermined target mass. It is further important that the filling process can be performed as rapidly as possible.
Known from the existing state of the art are dosage-dispensing methods which are based on a volumetric measurement of the amount of substance delivered. For a substance of density ρ, a valve with a variable aperture cross-section A and, associated with these parameters, a resultant outflow velocity u of the substance, the mass mz of the substance in the target container is obtained from the equation
In particular the outflow velocity u is subject to many factors such as for example the cross-sectional area A of the valve aperture, the hydrostatic pressure resulting from the fill height h of the substance in the reservoir, and the rheological properties of the substance such as for example the grain size d of the powder. The rheological properties are often very complex and are subject to factors which are not known with any degree of precision. It is for example difficult to take the flow retardation into account which occurs in Bingham media or powders at the beginning of the flow movement. Particularly in the filling of pulverous substances, factors such as for example grain size, moisture content, and surface properties of the individual particles, have a major influence.
In U.S. Pat. No. 4,893,262 a controller system for the filling of containers is disclosed. The system is optimized in a process that extends over several filling cycles, wherein the mass flow is optimized from one cycle to the next and the filling time is adjusted until the dispensed mass agrees as accurately as possible with the predetermined target mass. This system is used primarily for the filling of large quantities where the requirements concerning accuracy are considerably lower than for the system of the present invention. The fact that the optimization extends over several cycles represents a further problem, as the predetermined target mass is actually attained with the required accuracy only after several trial cycles. The substance dispensed during these trial cycles cannot be used anymore, since it could have been contaminated in the process of dispensing and subsequent removal from the target container. This is a decisive disadvantage especially when the substances being dispensed are expensive.
A method and apparatus to dispense a small mass of particles accurately and reproducibly are disclosed in U.S. Pat. No. 6,987,228 B1. The apparatus includes a controller unit which serves to control the amount of energy imparted to a sieve which holds the particles to be dispensed. Energizing the sieve has the effect that a small quantity of the particles in the sieve will fall on a balance that is arranged below the sieve. Based on the weight measured by the balance, the controller unit controls the amount of energy that is applied to the sieve. The amount of energy being introduced can be controlled as a function of the amount of mass that remains to be dispensed, whereby the outflow rate of the particles can be varied. This arrangement has the problem that with the use of the sieve, only substances in powder form can be dispensed. For other free-flowing substances, particularly for liquids, this method is not suitable. Even when pulverous substances are being dispensed, there are drawbacks inherent in this method, as different sieves have to be used depending on the grain size of the substance. The essential disadvantage concerns the control of the energy supplied to the sieve as a function of the weighing signal. Because of the time delay in the response of the balance, the filling process would have to be performed at a slow enough speed to allow enough time for the balance to respond. As a result, the filling process would take a very long time.
A system to control the filling of containers is disclosed in U.S. Pat. No. 4,762,252. To determine the mass flow rate during the filling process, the change in the weight of the reservoir is measured. The mass flow rate that is determined in this way is compared to a desired flow rate. If the measured flow rate deviates too strongly from the desired flow rate, the mass flow rate is adjusted accordingly. The system described in this reference is suitable for dispensing about 25 to 50 kilograms per hour. In the filling of small quantities of the sizes required in the field of pharmaceuticals, a high level of accuracy is required and small measurement inaccuracies can have a significant influence on the fill mass. At the same time, the filling process should take as little time as possible.
It is therefore an object of the present invention to provide a method and an apparatus serving to accurately and reproducibly dispense a predefined small fill quantity of a free-flowing substance and having the attributes of being simple, fast and accurate.
This task is solved with a method and an apparatus having the features described in the disclosed embodiments or the claims.
According to the disclosed methods and with the disclosed embodiments of the apparatus, a target container is filled with a predetermined target mass mz of a free-flowing substance from a reservoir with the help of a dispensing device for the filling of measured doses of the substance into the target container. The dosage-dispensing device has a valve which allows a variable adjustment of the mass flow rate {dot over (m)} from the reservoir into the target container. The dosage-dispensing device further includes a means for measuring the elapsed time t from the beginning of the filling process, a balance for determining the mass m of the substance in the target container, and a controller unit with a valve control module for controlling the valve. The controller unit includes an adjustment module, wherein a desired mass flow rate {dot over (m)}★ stored in the adjustment module. If at a point in time t the mass flow rate {dot over (m)}(t) is smaller than the desired mass flow rate {dot over (m)}★, the flow rate {dot over (m)}★ is increased by a mass flow rate adjustment
d{dot over (m)}={dot over (m)}★−{dot over (m)}(t),
and if the mass flow rate {dot over (m)}(t) is larger than the desired mass flow rate {dot over (m)}★, the flow rate {dot over (m)}★ is decreased by a mass flow rate adjustment
d{dot over (m)}={dot over (m)}(t)−{dot over (m)}★.
It is advantageous if several different parameters enter into the determination of the desired mass flow rate {dot over (m)}★.
One of these parameters is the delay interval τ. The delay interval τ represents the time interval from the arrival of the mass on the balance until the mass value is indicated on the balance. The delay interval τ is determined through measurements. It has been found that in the majority of cases the delay interval τ depends on the characteristics of the balance. In addition, parameters of the ambient environment have an influence on the delay interval τ. For example, low-frequency vibrations and/or tremors will cause an increase of the delay interval τ. This means that the primary factors entering into the delay interval τ are the balance-specific measurement delay and environmental parameters. Physical properties of the substance to be dispensed play only a secondary part.
In order to ensure that the mass dispensed into the target container does not exceed the limit, one can let the delay interval τ enter into the determination of the desired mass flow rate {dot over (m)}★. The rule to be followed here is that the delay interval τ should be in inverse proportion to the desired mass flow rate {dot over (m)}★, i.e.
A further parameter that should enter into the determination of the desired mass flow rate is the mass tolerance mT. The mass tolerance mT defines the maximally allowable deviation by which the mass m that is in the target container at the end of the dosage-dispensing process may differ from the target mass mz. In other words, the mass m that is in the target container at the end of the dosage-dispensing process must lie within the interval
mz−mT<m<mz+mT.
If the mass tolerance mT is large, the desired outcome of the end mass m being within the given tolerance mT can be assured even in the case where a large mass flow rate {dot over (m)}★ is desired. On the other hand, if the mass tolerance mT is small, the desired mass flow rate {dot over (m)}★ needs to be selected small enough to ensure that the mass m ends up within the given tolerance mT. This leads to the requirement that the desired mass flow rate {dot over (m)}★ should be in proportion to the mass tolerance mT, i.e.
{dot over (m)}★∝mT.
If the end mass m of the substance dispensed into the target container exceeds the target mass mz by more than the mass tolerance mT, i.e. if
m>mz+mT,
this is considered an overshoot. Overshooting the tolerance mT needs to be strictly prevented, as the excess amount of substance delivered cannot be removed again from the target container without great difficulty. Also, in the process of removing the substance from the target container it is possible that contaminations will occur, a risk that needs to be avoided.
In order to safely avoid an overshooting of the mass received by the target container, the amount of mass that is dispensed into the target container during the delay interval τ should be smaller than the tolerance mT, i.e.
{dot over (m)}★·τ≦mT.
Consequently, the mass flow rate has to be smaller than or equal to the tolerance mT divided by the delay interval τ.
The maximum allowable value for the desired mass flow rate {dot over (m)}★ is therefore
By staying below this maximum value, one ensures that the mass m in the target container at the end of the filling process is within the predefined tolerance range.
As the delay interval τ can change during a filling process, depending on the environmental parameters, this change in the delay interval τ can be used to make an adjustment in the desired mass flow rate {dot over (m)}★.
The adjustment module compares the desired mass flow rate {dot over (m)}★ to the actual mass flow rate {dot over (m)}(t) and if a difference is found between the two quantities, the adjustment module adapts the actual mass flow rate {dot over (m)}(t) to the desired mass flow rate {dot over (m)}★. To ensure that the desired mass flow rate {dot over (m)}★ is maintained over the entire filling process, the adjustment module is used repeatedly. It is particularly advantageous to repeat the operation of the adjustment module after equal time intervals.
To prevent the system from becoming unstable, the mass flow rate {dot over (m)} should not be changed too rapidly. It is therefore advantageous to let the previous mass flow rate {dot over (m)}old enter into the calculation of the new mass flow rate {dot over (m)}new, in accordance with this equation:
{dot over (m)}new=(1−α){dot over (m)}old+α({dot over (m)}old−d{dot over (m)}).
The factor α is a weight factor which can take on any desired value between zero and one, i.e. αε(0,1). This has the effect that the mass flow rate {dot over (m)} will change more slowly.
One possibility to prevent an overshooting of the target mass mz is to calculate the actual mass {tilde over (m)}(t) that is present in the target container based on the current mass flow rate {dot over (m)}(t), the weighing signal m(t) measured by the balance, and the delay interval τ, using the relationship
{tilde over (m)}(t)=m(t)+τ·{dot over (m)}(t).
Based on the calculated amount {tilde over (m)}(t) for the actual mass that is present in the target container, the closing of the valve can be started at the right time and an overshooting of the target mass mz can thereby be avoided.
The risk of overshooting the target mass mz is further reduced, if the desired mass flow rate {dot over (m)}★, and consequently also the actual mass flow rate {dot over (m)}(t), is lowered towards the end of the filling process.
The disclosed methods and the disclosed embodiments of the apparatus find application in particular in the filling of pulverous or liquid substances. The free-flowing substances normally have complex rheological properties and are in most cases of a non-Newtonian nature. The desired target mass is typically in a range between 0.5 mg and 5000 mg. However, it is also possible to dispense smaller or larger dosage quantities with this method.
According to an advantageous embodiment, the valve has an outlet orifice of circular cross-section and a shutter element, wherein the outlet orifice and the shutter element are arranged on a common axis. The shutter element, which has the mobility relative to the housing to rotate about the common axis and to slide in translatory movement along the common axis, can be driven out of, and retracted back into, the outlet orifice. The shutter element has a cylindrical shutter portion and an outlet passage portion, so that by a translatory movement of the shutter element equal to the length L of the latter the valve can be opened and closed. The outlet passage portion of the shutter element is designed so that the substance can flow through the outlet passage portion if the translatory displacement is larger than a minimum translatory displacement Lmin, and smaller than a maximum translatory displacement Lmax. If the translatory displacement is smaller than the minimum translatory displacement Lmin, or larger than the maximum translatory displacement Lmax, the outlet orifice is closed off by the cylindrical shutter portion and the substance cannot pass through the outlet passage portion.
The mass flow rate {dot over (m)} is directly correlated with the translatory displacement L of the shutter element.
To arrive at the actual mass flow rate {dot over (m)} one can also make use of the weighing signal m(t) by determining the first time derivative of the weighing signal
The time difference Δt>0 can be chosen arbitrarily. However, attention should be given to choosing Δt sufficiently large, so that statistical fluctuations of the weighing signal are smoothed out and the values for {dot over (m)} will not excessively fluctuate as a result. Excessive fluctuations can lead to instability of the dosage-dispensing process. In determining the mass flow rate {dot over (m)}(t), it is preferred to use discrete, uniform time intervals Δt. From the time t1 when the valve is opened, the mass flow rate {dot over (m)}(ti) at the times ti which follow each other in uniform steps Δt according to the sequence
ti=t1+iΔt
is determined with the help of the following equation:
Ideally, n is a positive integer between 2 and 10. If a larger value for n is used, the mass flow rate {dot over (m)}(ti) is determined over a larger time interval, whereby on the one hand statistical fluctuations of the weighing signal are smoothed out. On the other hand, if a large time interval is used, the value of {dot over (m)}(ti) is relatively slow to respond, so that a change in the mass flow rate will be detected relatively late.
Based on the calculated mass flow rate {dot over (m)}(t), the deviation
d{dot over (m)}={dot over (m)}★−{dot over (m)}(t)
from the desired mass flow rate {dot over (m)}★ can be determined, and the mass flow rate {dot over (m)}(t) can be adjusted accordingly.
It is further of advantage that the valve is opened and closed by moving the shutter element in translatory steps of equal magnitude ΔL.
Ideally, in the filling process a determination is made of the minimum translatory displacement Lmin, which allows the substance to flow and/or of the maximum translatory displacement Lmax. It is further practical to determine the actual mass that is present at the end of the filling process. These parameters can be stored and used by the controller unit in subsequent fill cycles. Thus, these parameters need to be determined only once, and subsequent fill cycles can be performed faster. The parameter values of preceding fill cycles can be stored in a memory unit, in particular an RFID (Radio Frequency Identification) tag and used in later fill cycles. It is particularly advantages to affix the RFID tags to the pertinent reservoir, as this will ensure a direct connection between the substance in the reservoir and the data stored in the RFID tag. However, one could also use other memory storage media.
Ideally, in a procedure that is used only in the first filling cycle, the valve is opened in translatory stepwise movements ΔL of the shutter element until the substance begins to flow, whereby the minimum translatory displacement Lmin, is defined. To determine the maximum translatory displacement Lmax, the shutter element is first opened by an amount Lmin+ΔL, and then the movement is continued in translatory steps of equal magnitude ΔL until the substance ceases to flow, whereby the maximum translatory displacement Lmax is defined. Once the minimum translatory displacement Lmin, and the maximum translatory displacement Lmax are known, the shutter element can be opened and closed in any way desired.
Advantageously, the outlet passage area of the shutter element has a variable aperture cross-section A. Thus, the mass flow rate {dot over (m)} of the stream of substance passing through the valve is directly correlated to the position of the shutter element of the valve. Ideally, the length L of the translatory displacement of the shutter element is directly correlated to the aperture cross-section of the valve, i.e. A=A(L). Depending on the design of the valve, there is a direct proportional relationship between the translatory displacement L of the shutter element, the aperture cross-section A, and the mass flow rate {dot over (m)}:
ΔL∝ΔA∝{dot over (m)}3.
Based on the geometric design of the shutter element, the dependence of the flow rate on the translatory displacement L is expressed by a cubic function and has been determined experimentally. A change of the geometry of the shutter element would also lead to a change in the relationship between the translatory displacement L and the mass flow rate {dot over (m)}.
However, a direct proportionality of this kind is normally not achievable in practical reality, as material properties such as for example grain size, delayed start of the flow movement, or similar factors act against a direct proportionality. However, it can normally be taken as a rule that with a larger aperture cross-section there will be a larger mass flow rate.
It is advantageous to use a shutter element which can be set into rotation with a variable angular velocity ω, wherein the angular velocity ω correlates directly to the rate {dot over (m)} of the mass flow through the valve.
As a further advantage, the valve is equipped with an impact mechanism delivering a tapping action of a variable tapping frequency F against the already open valve. In this case the tapping frequency F correlates directly to the mass flow rate {dot over (m)} through the valve, and an increase of the tapping frequency F leads to a larger mass flow rate {dot over (m)}. The tapping strikes can be directed parallel as well as transverse to the axial direction of the shutter element.
Furthermore, the taps can strike against the shutter element of the valve and/or against the housing of the valve.
With both the rotary movement and with the tapping action one gains the benefit, that a clogging of the valve and/or the formation of powder bridges can thereby be counteracted. In this way, the free-flowing property of the powder can be preserved or even enhanced.
The controller unit can be realized in part or in its entirety as a computer-based system.
The method and the apparatus for the filling of target containers are hereinafter described through examples which are illustrated schematically in the drawings, wherein
From two mass values m(t−Δt) and m(t) which were determined at two consecutive points in time t and t+Δt, the mass flow rate can be determined as
The aim in this process is to fill the target container 100 at a desired mass flow rate {dot over (m)}★ until the desired amount of mass is present in the target container. Ideally, the desired mass flow rate {dot over (m)}★ meets the condition
wherein mT represents the tolerance value for the discrepancy between the desired target weight and the actual end weight, and τ represents the delay interval of the balance. The delay interval τ is a balance-specific parameter which is independent of the physical properties of the substance to be dispensed. The delay interval τ can be determined prior to the first fill cycle and stored in the controller unit 600.
The delay interval time τ depends on the technical characteristics of the balance 500 and on the parameters of the ambient environment. The ambient parameters can change over the course of a filling process, and this can also cause a change of the delay interval τ. This change of the delay interval τ can be determined continuously, and the desired mass flow rate {dot over (m)}★ can be adapted in response to the change of the delay interval τ.
The calculated mass flow rate {dot over (m)}(t) is passed on to an adjustment module 620 where the mass flow rate {dot over (m)}(t) that has been determined from the measurement values is compared to the desired mass flow rate {dot over (m)}★. If the calculated mass flow rate {dot over (m)}(t) is found to be smaller than the desired mass flow rate {dot over (m)}★, the mass flow rate {dot over (m)} is increased by d{dot over (m)}, and if the calculated mass flow rate {dot over (m)}(t) is found to be larger than the desired mass flow rate {dot over (m)}★, the mass flow rate {dot over (m)} is decreased by d{dot over (m)}. After the adjustment, the actual mass flow rate {dot over (m)}(t) should agree with the desired mass flow rate {dot over (m)}★. The adjustment module 620 sends the signal for the adjustment of the mass flow rate to the valve 310. The determination of the actual mass flow rate is performed repeatedly during the fill cycle and, if necessary, the mass flow rate {dot over (m)}(t) is adjusted. The determination of the actual mass flow rate and/or the adjustment of the mass flow rate can be performed in equal time intervals.
Between the shutter element 313 and the valve housing 311 there is a hollow space which serves as a reservoir 200 for the substance to be dispensed. A translatory movement 340 of the shutter element 313 opens the way so that the substance to be dispensed can pass from the reservoir 200 by way of the outlet passage portion 315 of the shutter element 313 and through the outlet orifice 312 into the target container 100.
The valve 310 includes a memory unit 320 to store data. In this memory unit 320 it is possible to store for example material properties of the substance to be dispensed, flow parameters from preceding filling processes, and/or balance-specific parameters such as for example the delay interval τ. The memory unit 320 is arranged on or in the valve housing 311.
In a comparison of the actual mass flow rate {dot over (m)}(t) against the desired mass flow rate {dot over (m)}★, if the actual mass flow rate {dot over (m)}(t) is found to be larger than the desired mass flow rate {dot over (m)}★, the actual mass flow rate {dot over (m)}(t) is lowered by reducing the aperture cross-section A of the valve 310. On the other hand, if the comparison shows that the actual mass flow rate {dot over (m)}(t) is smaller than the desired mass flow rate {dot over (m)}★, the actual mass flow rate {dot over (m)}(t) is raised by increasing the aperture cross-section A of the valve 310.
Towards the end of the filling process the aperture cross-section A of the valve is reduced, whereby the mass flow rate {dot over (m)}(t) is lowered. In this way, the mass m(t) in the target container 100 can approach the target mass mz slowly, whereby an overshooting of the target mass mz is prevented.
Although embodiments of the invention have been described by presenting specific exemplary embodiments, it is obvious that numerous further variants could be created based on a knowledge of the disclosed embodiments, for example by combining the features of the individual examples of embodiments with each other and/or by interchanging individual functional units between the embodiments.
| Number | Date | Country | Kind |
|---|---|---|---|
| 08162902 | Aug 2008 | EP | regional |
This application is a continuation under 35 USC §120 of PCT/EP2009/060524, filed 13 Aug. 2009, which is in turn entitled to benefit of a right of priority under 35 USC §119 from European patent application 08 16 2902.4, filed 25 Aug. 2008. The content of each of the applications is incorporated by reference as if fully recited herein.
| Number | Name | Date | Kind |
|---|---|---|---|
| 4762252 | Hyer et al. | Aug 1988 | A |
| 4893262 | Kalata | Jan 1990 | A |
| 6987228 | MacMichael et al. | Jan 2006 | B1 |
| 20090293986 | Blochlinger | Dec 2009 | A1 |
| Number | Date | Country |
|---|---|---|
| 3910028 | Oct 1989 | DE |
| 0133176 | May 2001 | WO |
| Number | Date | Country | |
|---|---|---|---|
| 20110172934 A1 | Jul 2011 | US |
| Number | Date | Country | |
|---|---|---|---|
| Parent | PCT/EP2009/060524 | Aug 2009 | US |
| Child | 13032953 | US |
