Force-measuring device using electromagnetic force compensation with constant power dissipation

This application is entitled to benefit of a right of priority under 35 USC §119 from European patent application 09 16 0710.1, which was filed on 20 May 2009. The content of the European application is incorporated by reference as if fully recited herein.

The embodiments disclosed herein relate to a force-measuring device, in particular a balance, which relies on the principle of electromagnetic force-compensation.

Under the electromagnetic force compensation measurement principle, an opposing force, also referred to as compensation force, is generated which counteracts the force that is to be measured by the force-measuring device. As a result of this compensation, the movable parts of the force-measuring device which serve to receive and transfer the force are regulated to maintain a set position. Accordingly, in the case of a balance the weight of the weighing object represents the force to be measured, and the movable parts include components such as the balance pan, levers, rods, or the weighing cell.

In many cases, the compensation force is generated by an electromagnetic coil which is arranged so that it can move within an air gap of a magnet system and which carries an electric current that is sent through it. This current, referred to as compensation current, represents a measure for the compensation force that is being generated and thus represents a measure for the force that is acting on the force-measuring device. By measuring this compensation current, a corresponding measuring signal is obtained which is analyzed in a signal-processing unit and converted into results that are delivered to an indicator device.

Force-measuring devices based on electro-magnetic force compensation have the disadvantage that the power which is generated in the coil by the compensation current depends on the magnitude of the current flowing and thus on the amount of force acting at any given time. Therefore, with measurements of different forces or loads following each other, different levels of power are generated and released in the form of heat.

As a consequence of the varying amounts of heat being released and the temperature changes that occur as a result, the zero point and the span of the measurement range can become unstable. These influence factors should therefore as much as possible be kept constant, particularly in force-measuring devices which have to meet exacting requirements regarding the invariability of their measurement accuracy.

The known state of the art offers different methods of achieving a power dissipation of the coil that is constant and independent of the weighing load. It is possible to use for example an additional power-consuming device which converts additional electrical energy into heat. As a result, the total amount of power consisting of the sum of the coil power and the additionally dissipated power of the power-consuming device remains largely constant. In this arrangement, the additional power-consuming device should as much as possible be neutral in its behavior in regard to force effects, so as not to influence the compensation force being generated.

For example, in DE 28 19 451 a variety of force-neutral power-consuming devices are disclosed in the form of a power transistor, a semiconductor resistor, or a coil with bifilar windings where the two coil parts are controlled so that their currents are anti-parallel, i.e. of equal magnitude but opposite direction. In a coil with bifilar windings and equal anti-parallel current flows, two forces are produced which mutually neutralize each other, so that the result is a power-consuming device which is force-neutral in its overall effect. Depending on its implementation, this concept has the following drawbacks:

The location where the non-productive power is generated is not geometrically identical with the location where the productive power is generated. Consequently, the temperature distribution in the force-measuring device still depends on the compensation force that is to be measured. The non-productive power is released in a coil part with anti-parallel current flows that is located within the compensation coil. With this arrangement, the degree of efficiency of the force-measuring device is reduced because only part of the coil is usable for the generation of the force even at the maximum level of compensation force.

As a further way of implementing a force-neutral power-consuming device, the concept of adding an alternating current to the compensation current is disclosed in DE 31 36 171. Thus, the resultant average compensation force of the coil remains essentially unchanged, while an additional power consumption is achieved through the AC component of the current. By alternating the direction of the current, the differences in the two windings are averaged out. This form of heat generation has the disadvantage that extensive measures have to be taken to prevent the AC component from causing a change of the average compensation current.

This disadvantage can be circumvented by using a coil with two separate windings acting in opposite directions which are energized alternatingly. In this arrangement a current flowing through the first winding generates a first force, and a current flowing through the second winding generates a second force. As an average, the two forces will produce a resultant force which represents the effectively acting compensation force.

For the control of the two currents, a switch-over device is disclosed in CH 634 654 which serves to direct the compensation current through the first winding during a first time interval and through the second winding during a subsequent second time interval. This has the result that the total compensation current over the two time intervals remains to a large extent constant. However, the periodic switching of the current flows generates strong alternating forces in the coil and thus causes a considerable amount of audible noise.

If an effort is made to mitigate the negative effects of the abrupt periodic alternations of the current by means of filter elements or smoothing capacitors, the power dissipation in the coil will again strongly depend on the compensation force generated at any particular moment and thus on the weighing load that is present at any given time.

It is therefore an object to propose a force-measuring device which is based on the principle of electromagnetic force compensation and which is distinguished by the fact that a simple, cost-effective design and operation of the force-measuring device can be achieved while the force-measuring device at the same time meets stringent requirements in regard to its measurement accuracy and efficiency.

A further object is to propose a force-measuring device which meets the aforementioned requirements and which can be operated without making an irritating noise.

A further aim is to achieve an optimal utilization of the space available for the windings. This is the space which is allocated to the part of the coil system that carries the windings. It is in essence delimited by the dimensions of the permanent magnet, in particular the magnet gap in which the coil system is arranged.

These tasks are solved with a force-measuring device and with a method possessing the features that are presented in the independent claims. Advantageous embodiments are described in the further, dependent claims.

The disclosed embodiments concern a force-measuring device, in particular a balance, which is based on the principle of electromagnetic force-compensation, with a movable electric coil which is arranged in a magnet system and has at least two windings, further with a current supply device which has at least two partial current sources, wherein each of said windings has one of said partial current sources assigned to it, further with a controlling device and/or regulating device which controls and/or regulates the supply of current to the windings by means of the partial current sources in such a way that, dependent on the force acting on the force-measuring device, a current is sent through each of the windings in such a way that the sum of the at least two electromagnetic forces which are thereby generated forms the compensation force and that at the same time the power dissipated by the coil always takes on a given predetermined value. By supplying the windings with current individually and simultaneously in this manner, the rate of heat generation in the coil can be controlled in a simple and efficient way without periodic alternation of the direction of current flow. Furthermore, with a current supply essentially in the form of direct current, the problem of audible noise does not exist.

The given predetermined value of the power needs to be larger than the power required at each moment to produce the compensation force, or else the required compensation force cannot be generated. Except for this limitation, the given predetermined value of the power can in principle be freely selected. For example, this value can be specified dependent on the load that is acting on the force-measuring device, so as to achieve an optimal temperature distribution. However, the given predetermined value of the power can also be defined in accordance with special predetermined conditions, as is described in the examples of embodiments.

The disclosed embodiments further concern a suitable device for controlling a force-measuring device, in particular a balance, which is based on the principle of electromagnetic force-compensation, with a movable electric coil which is arranged in a magnet system and has at least two windings, and using the following method in order to supply said windings with current: dependent on the force acting on the force-measuring device, a current is sent through each of the windings in such a way that the electromagnetic forces which are thereby generated form in their overall outward effect the compensation force. In this arrangement, the at least two currents are sent through the windings and the respective strengths of the currents are controlled and/or regulated by taking into account the total power generated in the coil, in such a way that the magnitude of said power remains constant, independent of the resultant compensation force.

Due to the coil having more than one part, an additional degree of freedom is gained for managing the current, which offers the possibility to control the overall power dissipated in the coil, independent of the required compensation force. In other words, it becomes possible to control and/or regulate the force generation and the power in the coil independently of each other in such a way that they are not mutually influencing each other.

This concept has the important advantage that it allows higher overall levels of power to be achieved. Consequently, the measurement range can be significantly increased while keeping the same lever ratio. For example, a measurement range of 0 to 400 grams can be expanded to a measurement range of 0 grams to 1 kilogram. Alternatively, or in addition, it is also possible to use a smaller lever ratio.

The embodiment proves especially advantageous if the wires of the individual windings have been wound together (i.e., in the case of two wires, in the form of a bifilar winding). With this kind of winding arrangement, the temperature generated by the power dissipation in the components of the force-measuring device and the local distribution of said temperature are kept optimally constant, whereby a high level of measurement accuracy and stability of the force-measuring device are achieved.

In an advantageous embodiment, the given predetermined value is constant. Under this concept, the given predetermined value can either be constant over the entire load range of the force-measuring device, or it could also be constant only over a part of the load range. It is conceivable, for example, that within a range from zero load to 80% of capacity the given predetermined value is prescribed as being constant and within the range from 80% to 100% of capacity the given predetermined value is prescribed according to whatever power is required at the moment in order to generate the compensation force. As a result of this at least partial independence of the power dissipation on the load, a high stability of the measurements is achieved.

In a further advantageous embodiment the given predetermined value represents a minimum relative to different power values which are obtained for a load range within the measurement capability of the force-measuring device, in particular for the entire load range of which the device is capable. This allows the temperature difference between the coil and the ambient temperature to be kept at as low a level as possible, because it avoids a generation of excess heat.

With preference, the force-measuring device includes a current-measuring device which measures the force-compensation current or -currents so that dependent on the measured values the currents through the windings can be controlled and/or regulated.

According to a further embodiment the force-measuring device includes a power estimator which serves to determine the power in the coil that is required at any given moment for generating the compensation force. It is further preferred to calculate and/or estimate the power of the coil based on the sum of the partial currents, using in particular the formula:

P 0=R×(I1+I2) 2 ,

wherein R represents the total resistance of the coil and I 1 , I 2 represent the partial currents of the coil windings (W 1 , W 2 ) that run parallel to each other. The total resistance R of the coil based on the parallel arrangement of the two coil windings is obtained with the formula:

R = 1 ( 1 R ⁢ ⁢ 1 + 1 R ⁢ ⁢ 2 )

Next, the given predetermined value Ptg for the control and/or regulation is derived from the momentary power P 0 . In this way, the control can be established in a simple and efficient manner.

It is particularly advantageous if the given predetermined value Ptg represents the power of the coil at maximum compensation force.

Preferably, the force-measuring device is equipped with a suitable control- and/or regulation device which allows the at least two currents to be generated as largely continuous currents. This control- and/or regulation device can be assembled from a plurality of discrete circuit elements or from integrated circuits, or it can be implemented in the form of program parts in a microprocessor. Especially with a high degree of integration, it is possible to realize a particularly cost-effective design of the control- and/or regulation device.

In a further embodiment the force-measuring device has a control quantity generator for the purpose of calculating a control quantity which serves to control and/or regulate the partial current sources and which, being constituted in particular as a nonlinear function, depends on the power P 0 required to generate the compensation force at any point in time.

Preferably, the force-measuring device includes a power distributor, in particular a digital potentiometer, which serves to control and/or regulate the partial current sources in accordance with the control quantity. This allows a cost-effective design to be achieved with standardized components.

In a further preferred embodiment the control quantity generator is configured in such a way that the control quantity x is calculated according to the formula

x=a−b√{square root over (P*)}+c √{square root over (( P*− 1))},

wherein a, b and c represent given constants and P* represents the normalized power of the coil, which is calculated as

P * = Ptot P ⁢ ⁢ 0

in terms of the full-load power Ptot and the momentary power P 0 of the coil. The momentary power P 0 in this context is the amount of power needed to generate the compensation force which is required at any point in time.

In a further preferred embodiment, each of the partial current sources is connected directly to a respective coil winding to supply the latter with current. This allows the currents in the respective coil windings to be controlled and/or regulated directly and thus in a particularly efficient manner.

Furthermore, the partial current source is advantageously equipped to supply the at least one coil winding with a direct current in accordance with a given offset voltage.

In an advantageous embodiment, the force-generating coil is divided into a plurality of parts, preferably two parts, with separate, parallel controls, wherein the windings of the coil parts preferably have the same number of turns. This means that each of the windings is capable of producing a contribution to the compensation force, and if the windings are operated in parallel, the sum of the partial currents I=ΣI i constitutes the compensation current required to generate the compensation force. With this kind of force-generating coil consisting of more than one part, preferably two parts, each of the parts is supplied separately from a partial current source, a so-called current booster. The arrangement further includes a current-measuring device which measures the compensation current required to hold the moving parts of the force-measuring device in balance. Using a power estimator which is based on this compensation current, the compensation force required to bring the force-measuring device into balance is generated by appropriately controlling the coil parts with the help of a power distributor in such a way that the power of the entire coil is held constant at a minimal value independent of the compensation force required any point in time.

In an advantageous embodiment, the at least two currents are controlled and/or regulated simultaneously in accordance with a control quantity that is dependent on the momentary power of the compensation current of the coil. This simplified concept opens the possibility to reduce the complexity of the control and thus build a particularly cost-effective and stable circuit.

In a further advantageous embodiment, the total power generated in the coil at the maximal compensation current is:

Ptot = ∑ R i × I i 2

wherein R i stands for the resistance values and I i stands for the partial currents of the coil windings that run parallel to each other.

As the compensation current Icp represents the actual weighing signal w, this current value needs to be measured and, if necessary, converted and evaluated. Based on the value for Icp, the coil power P 0 which has to be released in order to generate the compensation force as required at the moment is determined by the elementary equation

P 0=R×I cp 2

wherein R represents the ohmic equivalent impedance of the entire coil and Icp represents the momentary compensation current.

In a particularly advantageous embodiment, the current strengths

I1= x ×I+Ioffset1

I2=(1 −x )×I−Ioffset2

in a coil with two coil parts are controlled in accordance with at least one nonlinear function that is dependent on the total power.

Preferably, this function is determined in essence by the following formula:

x=a−b√{square root over (P*)}+c √{square root over (( P*− 1))},

wherein a, b and c are constants and P*=Ptot/P 0 represents the ratio between the full-load power Ptot of the coil at the maximal compensation current and the power P 0 required at any point in time to generate the compensation force. What is accomplished with this profile of the function is that the total coil power will have an optimal degree of independence on the weighing load. It is also conceivable to use other suitably designed function profiles that have similar characteristics, for example composites of simple or higher-order polynomials or of exponential functions.

This calculation is preferably performed by a control device and/or regulation device which is arranged in the force-measuring device. For example, these operations can be performed through an analog process by means of simple circuits or through a digital process by means of a small number of computational steps in a microprocessor.

In a particularly advantageous embodiment, an offset current is superimposed on each of the two coil currents. These currents have the purpose that a constant level of power is generated in the coil even when the compensation current is small. In a preferred implementation of this concept the control- and/or regulation device has a partial current source dedicated to each coil winding and connected to the latter in order to supply the winding with a direct current in accordance with a given offset voltage.

The device according to this embodiment always maintains the ability to bring the movable part of the force-measuring device into a balanced state, regardless of whether or not the compensation current changes polarity between its minimum and maximum values.

In a further embodiment, the control- and/or regulation device includes at least two switching elements that serve to modulate the partial currents into a substantially pulse-shaped form, wherein in the time profile of the partial currents a time interval can be prescribed during which the partial currents are flowing simultaneously. Using the switching elements it is possible to generate an additional amount of heat with simple and cost-effective means.

The basis for the time profile is a periodic process characterized by the length of a period. In this process, a period is defined by the simultaneous periodic switching of the switching elements whereby the switching of the two partial currents is performed. With this simultaneous actuation of the switching elements there is no additional heat being generated, so that this operating mode reflects the state of the art. Furthermore, for this operating mode, an effective time interval t* is defined during which the first partial current flows through the first winding. From the effective time interval t* and the period length ts, the weighing signal w can be determined in accordance with the ratio

w = t * ts

Thus, the weighing signal is connected directly to the applied force.

According to the disclosed embodiments, in order to generate an additional amount of heat the first control signal is switched with a delay relative to the end of the effective time interval t*. The length of the delay is referred to as a first time interval tp 1 , i.e. the switching occurs at the time t*+tp 1 . Likewise, the second control signal is not being switched at the beginning or at the end of a period ts, but delayed by a second time interval tp 2 , i.e. at the time ts+tp 2 . These events repeat themselves periodically at the times 0, ts, 2ts, 3ts, etc. Accordingly, there are within each period ts a first time interval tp 1 and a second time interval tp 2 during which the control signals overlap. During these time intervals, the first partial current and the second partial current are thus flowing simultaneously through the first and second windings, respectively. This has the consequence that during these time intervals, heat is generated in a force-neutral way in the windings.

Preferably, the first time interval tp 1 and the second time interval tp 2 are selected to be identical, so that there is symmetry between the first control signal SWI 1 and the second control signal SWI 2 . This symmetry is achieved through a largely identical choice for the two delay values tp 1 and tp 2 in the form of a common delay value tp.

With preference, the common delay value tp is calculated according to the following formula:

tp = ts × ( - 1 2 + 1 4 + w ⁡ ( 1 - w ) )

wherein ts represents the period length and w represents the weight signal. Through the value w, the delay value tp thus depends on the applied force L.

In an advantageous embodiment, the control- and/or regulation device includes at least one delay element which is connected to at least one switching element for the purpose of controlling the switching element by means of a control signal, wherein the control signal depends on a delay value which corresponds to the time interval. This provides a simple way to generate the control signals.

In a further advantageous embodiment, the control- and/or regulation device includes an arithmetic unit which allows the delay value to be calculated dependent on the applied force, wherein the arithmetic unit is connected to the delay element so as to transmit the calculated delay value to the delay element. This arrangement offers a simple way to incorporate complex relationships in the regulation.

In a further preferred embodiment, the force-measuring device includes a measuring unit which is connected to the arithmetic unit for the purpose of transmitting to the arithmetic unit a measuring signal dependent on the applied force, from which the load-dependent delay value can be calculated in the arithmetic unit.

The embodiment can also be used in designs with coils that have more than two coil windings. For example in a coil with three coil windings, two of the windings can generate a force in one direction while the remaining winding generates a force in the opposite direction. In the case of coils with four coil windings, the forces could be combined in a similar way.

Details of the method and the device that embody the inventive concept will be apparent from the description of the embodiments which are presented in a very schematic way in the drawings, wherein:

FIG. 1 depicts a balance 1 which, in regard to its design, may serve as an example, with a symbolically indicated load L, weighing pan 2 , and indicator unit 3 ;

FIG. 2 illustrates the principle of a force-measuring cell 10 which is based on the concept of electromagnetic force-compensation, shown in a cross-sectional view, with a coil 53 arranged to move in the air gap 51 of a magnet system 50 ;

FIG. 3 depicts, as an example, a schematic block diagram of the control- and/or regulation device CU, a current-measuring device M, signal-processing stages A/D and DSP, a schematically drawn magnet system 50 and a coil system 53 in accordance with FIG. 2 , which has two coil windings W 1 and W 2 ;

FIG. 4 shows: a) an example for the shape of the function x(P 0 ), i.e. the control variable x as a function of the power P 0 which is required at any point in time to generate the compensation force; b) the coil voltage U A1 and U A2 associated with the example of FIG. 4 a ; c) the currents I A1 and I A2 associated with the example of FIG. 4 a ; and d) the associated total power Ptot of the coil system;

FIG. 5 represents a detail of the block diagram of FIG. 3 , simplified and meant as an example, with a power distributor PD in the form of a digital potentiometer and, downstream of the latter, two partial current sources PB 1 and PB 2 ;

FIG. 6 represents a strongly simplified block diagram of a further embodiment with switching elements S 1 , S 2 , which are controlled, respectively, by control signals SWI 1 and SWI 2 in order to supply pulse-shaped currents to the associated coils;

FIG. 7 represents, in strongly simplified form, the time graphs of the control signals SWI 1 and SWI 2 that are indicated in FIG. 6 ;

FIG. 8 represents an example of an electronic circuit implementation of the block schematic of FIG. 6 , wherein the switching elements S 1 and S 2 are configured as switching transistors which are controlled by a single common control signal SWI; and

FIG. 9 represents, in strongly simplified form, the time graphs of the control signals SWI, SWI 1 and SWI 2 which are indicated in FIG. 8 .

FIG. 1 shows an example of a force-measuring device in the form of a balance 1 with a symbolically indicated load L. In the case of a balance, the force L that is to be measured is typically the weight or weighing load of the weighing object to be measured, which acts on the weighing pan 2 . The objective for the balance 1 is to arrive at an accurate and stable output value for the applied force L which is displayed on the indicator unit 3 , for example a digital liquid crystal display.

In a strongly simplified cross-sectional drawing, FIG. 2 illustrates a force-measuring cell 10 which is based on the principle of electromagnetic force-compensation and suitable for use in the field of weighing technology. The force-measuring cell 10 includes a force-transmitting mechanism with a parallel-guiding linkage having a stationary part 42 and a vertically displaceable part 43 which are linked to the ends of a pair of guide members 44 through flexures 45 . The vertically displaceable part 43 carries a weighing pan in the form of a cantilevered extension 41 which serves to receive a load L (indicated schematically) which is to be measured. The normal component of the force produced by a load L is transmitted from the vertically displaceable part 43 through a coupling element 49 to the short lever arm 48 of the lever 46 . The lever 46 is supported by means of a flexure fulcrum 47 on a portion of the stationary part 42 . The force-measuring cell further includes a cup-shaped magnet system 50 which contains an air gap and is attached to the stationary part 42 with a fixed connection. Arranged in the air gap 51 is a coil 53 which is connected to the longer lever arm of the lever 46 . The coil 53 carries an electrical current, the so-called compensation current Icp, whose magnitude depends of the force that is acting on the lever 46 . The position of the lever 46 is measured by a position-measuring device 54 which is connected to a control- and/or regulation device 60 . The control and/or regulation device 60 controls and/or regulates the compensation current Icp based on the measuring signals received from the position-measuring device 54 in such a way that the lever 46 is always held in the same position or is returned to the same position after a change in the load has occurred.

According to this measurement principle, the electromagnetic coil produces a force which acts on the lever, the so-called compensation force K which opposes the applied force L. The compensation current Icp generating the compensation force K represents a measure for the applied force L that is to be measured. By determining the magnitude of this compensation current Icp, followed by appropriate signal processing, a weighing signal w is generated which is transmitted to the indicator unit of the balance where it is displayed as the corresponding measurement value. However, the weighing signal w can also be passed on to a further processing unit, for example a lead computer, a system controller or a remotely located controller unit.

The electromagnetic coil 53 consists of one or more windings of an electrically conductive and insulated coil wire. In the simplified representation of FIG. 2 , only a few turns of the coil wire are indicated schematically. Typically, however, the coil 53 has windings with a large number of turns, for example several hundred.

Depending on the variable magnitude of the applied force L, a compensation current Icp varying in a corresponding manner is sent through the coil 53 . As a result a load-dependent momentary power P 0 is generated in the coil 53 which accordingly acts as a heat source. This heat source causes a corresponding temperature change of the affected components, in particular the coil 53 and the magnet system 50 . By using a load-independent amount of power, the embodiments aim to keep the additional temperature increase and its distribution in the force-measuring device at the most constant and lowest level possible.

As an example, FIG. 3 shows a schematic block diagram of a control- and/or regulation device CU, a current-measuring device M, a signal-processing unit in the form of an analog/digital converter A/D and a digital signal-processing unit DSP, a schematically drawn magnet system 50 with a coil 53 in accordance with FIG. 2 , which has two coil windings W 1 and W 2 . The cup-shaped magnet system 50 generates in its air gap a magnetic field B which surrounds the first coil winding W 1 and the second coil winding W 2 . For better clarity, the magnet system 50 is shown with its cylindrical shell cut open.

Each of the coil windings W 1 and W 2 of the coil 53 is indicated symbolically by a single turn, while in reality the coil windings typically have a large number of turns. The first coil winding W 1 is indicated as a solid line and the second coil winding W 2 as a broken line.

The measuring signal U PS generated by the position-measuring device is sent to a position control device PCL where it is evaluated and, if necessary, processed into a suitable form to be transmitted. The output of the position control device PCL is connected to a first input of a power distributor PD, so that the position regulator output signal U U of the position control device PCL is available as the first input variable for the power distributor PD. In this arrangement, the position regulator output signal U U corresponds to the measurement signal U PS received as input from the position-measuring device, so that the power distributor PD receives an input quantity which depends on the force L acting on the force-measuring device.

The power distributor PD is connected at its output side to two partial current sources PB 1 and PB 2 , so-called current boosters, so that the control voltages U H1 and U H2 generated by the power distributor PD can be sent independently of each other to the partial current sources PB 1 and PB 2 .

Using a first amplification factor K 1 , the first partial current source PB 1 generates a first coil voltage U A1 which is present at the output of the first partial current source PB 1 . The output terminal of the first partial current source PB 1 is connected to a first connector terminal of the first winding W 1 , and the second connector terminal of the first winding W 1 is connected to a ground potential. The line resistance of the first winding W 1 , the so-called first coil resistance R 1 is represented in the form of an equivalent circuit element as a resistor (dotted line). Now, if the first connector terminal of the winding W 1 is held at the first coil voltage U A1 , a first current I 1 which is generated by the partial current source PB 1 and whose magnitude depends on the first coil resistance R 1 will flow through the first winding W 1 to the ground potential.

In the same way as the first partial current source PB 1 and independently of the latter, the second partial current source PB 2 is connected to the second winding W 2 of the coil. Likewise, using a second amplification factor K 2 , a second coil voltage U A2 is generated by the second partial current source PB 2 , so that a second current I 2 which is generated by the partial current source PB 2 and whose magnitude depends on the second coil resistance R 2 will flow through the second winding W 2 to the ground potential.

The first partial current source PB 1 and the second partial current source PB 2 can be combined into a current supply device PB as indicated by a dash-dotted line.

The connection of the first winding W 1 to ground potential and the connection of the second winding W 2 to ground potential are combined into a common conductor line, so that the coil currents I 1 and I 2 are joined together in this part of the circuit and the sum of the coil currents I 1 and I 2 forms the compensation current Icp which flows to ground potential.

The magnitude of the summation current, i.e. the compensation current Icp, represents a measure for the compensation force K generated in the coil and thus a measure for the force L acting on the force-measuring device. Accordingly, the compensation current Icp is measured by means of a current-measuring device M, and the corresponding measuring signal is generated in the form of a weighing signal w which is converted into a digital format by means of an analog/digital converter. In a digital signal-processing unit DSP arranged downstream of the A/D converter the weighing signal w is subjected to further processing. The digital signal-processing unit DSP can include for example an amplifier, a low-pass filter, or a normalizer. At the output of the digital signal-processing unit DSP, the finished weighing signal is now available for further use, for example to be presented on a display or to be transmitted to a lead computer.

The output of the analog/digital converter A/D is connected to a power estimator PE, so that a digital weighing signal w corresponding to the compensation current Icp is presented at the input of the power estimator PE. The power estimator PE estimates the power P 0 that is needed in the coil at that moment for generating the compensation force in accordance with the following formula

P 0=R×I cp 2

wherein R represents the total resistance of the coil 53 and Icp represents the magnitude of the compensation current corresponding to the weighing signal w. As the two windings W 1 and W 2 are parallel to each other in the circuit, the total resistance R of the coil is found with the formula:

R = 1 ( 1 R ⁢ ⁢ 1 + 1 R ⁢ ⁢ 2 )

wherein R 1 represents the first coil resistance and R 2 the second coil resistance. In the case where the resistance values R 1 and R 2 are equal, R will be equal to one-half of the first coil resistance:

In addition, the power estimator PE performs the function of normalizing the momentary power P 0 in the coil in terms of a normalizing constant Ptot according to the relationship:

P * = Ptot P ⁢ ⁢ 0

wherein P* represents the normalized power of the coil and Ptot represents the total power of the coil at full load.

In the case of a force-measuring device, where the term “zero load” defines the operating condition where no force is acting on the force-measuring device while “full load” defines the operating condition where the applied force L equals a specified maximum of the measurable force, the full-load power Ptot can be calculated according to the formula:

P tot=R×I U 2

wherein IU represents the compensation current Icp at full load.

Accordingly, the normalized power P* of the coil lies between the values of infinity for zero load and 1 for full load.

The output of the power estimator PE is connected to an input of a control quantity generator XG, so that a value corresponding to the normalized power P* of the coil can be transmitted to the control quantity generator XG.

In the case at hand, the control quantity generator XG calculates from the input quantity, i.e. from the normalized power P*, a control quantity x according to the following equation:

x = 1 2 - 1 2 ⁢ P * + 1 2 ⁢ ( P * - 1 ) ,

Accordingly, the value of the control quantity x lies between the value of ½ at zero load and the value of 0 at full load.

The output of the control quantity generator XG is connected to a second input of the power distributor PD in order to transmit the calculated control quantity x to the power distributor PD, so that the latter receives the control quantity x as a second input quantity.

In the power distributor, the two aforementioned control voltages UH 1 and UH 2 are determined from the control quantity x and the position regulator output signal U U described above in accordance with the following equations:

U H1 =U U ×x+U of1

U H2 =U U ×(1− x )+ U of2

After the two control voltages U H1 and U H2 have been determined, the quantities through which the afore-described partial current sources PB 1 and PB 2 are controlled are now available. Consequently, this closes the loop to allow the power distributor PD to function in the manner described above.

Accordingly, this circuit arrangement represents two control loops, i.e. a first control loop for the regulation of the compensation force K and a second control loop for the distribution of the currents in such a way that the total power Ptot generated in the windings W 1 , W 2 is essentially constant over a range of different strengths of the compensation force K.

The individual components of the circuit can be appropriately combined with each other, for example as a control- and/or regulating unit CU (indicated by a broken line). Furthermore, the components of the circuit, in particular the power estimator PE and the control quantity generator XG, can be implemented as functional blocks in an integrated circuit or as program parts in a microprocessor. The control- and/or regulation device CU can thus be built using an analog or digital design or any kind of mixed configuration.

In a particularly advantageous embodiment, the first partial current source PB 1 generates the first current I 1 with an additional first offset voltage Uof 1 of a prescribed fixed magnitude. Analogously, the second partial current source PB 2 generates the second current with an additional second offset voltage Uof 2 of a prescribed fixed magnitude. Thus, the following currents I 1 and I 2 flow, respectively, through the windings W 1 and W 2 :

I ⁢ ⁢ 1 = K ⁢ ⁢ 1 R ⁢ ⁢ 1 × ( x × U U + U of ⁢ ⁢ 1 ) I ⁢ ⁢ 2 = K ⁢ ⁢ 2 R ⁢ ⁢ 2 × ( ( 1 - x ) × U U + U of ⁢ ⁢ 2 )

wherein Uof 1 and Uof 2 represent the offset voltages and the further symbols in the equations are the same as defined previously herein.

Preferably, the offset voltages Uof 1 and Uof 2 are determined according to the following equations in terms of the summation current IU at full load:

U of ⁢ ⁢ 1 = R ⁢ ⁢ 1 × IU ( K ⁢ ⁢ 1 × 2 ) and U of ⁢ ⁢ 2 = - R ⁢ ⁢ 2 × IU ( K ⁢ ⁢ 2 × 2 )

wherein IU represents a coil current which would occur under full load in a state-of-the-art force-measuring device.

Thus, as a result of the additional offset voltages Uof 1 and Uof 2 a direct current is superimposed on each of the currents I 1 and I 2 .

FIG. 4 a shows an example for the shape of the function F(P 0 ), i.e. the control variable x as a function of the power P 0 which is required in the coil at any point in time to generate the compensation force. In the graph, the control quantity x is represented as the ordinate and the power P 0 which is required at a given point in time to generate the compensation force is represented as the abscissa. The profile of the function F(P 0 ) corresponds to the previously described equation

x = 1 2 - 1 2 ⁢ P * + 1 2 ⁢ ( P * - 1 ) with P * = Ptot P ⁢ ⁢ 0

wherein P* represents the normalized power of the coil, the full load power Ptot represents the normalization constant, and P 0 represents the power required at any point in time.

At zero load, i.e. when P 0 =0, the control quantity x assumes its maximum value of x=½. At full load (P 0 =Ptot) the control quantity x assumes is minimum value of x=0. The profile of the function F(P 0 ) is nonlinear.

FIG. 4 b shows the profiles of the two coil voltages UA 1 and UA 2 which correspond to the control quantity x of FIG. 4 a ) and which are generated, respectively, by the partial current sources PB 1 and PB 2 . In this diagram, the coil voltage UA is represented as the ordinate axis and the power P 0 that is required at any point in time to generate the compensation force is represented as the abscissa. The profile of the first coil voltage U A1 is indicated by a solid line, and the profile of the second coil voltage U A2 is indicated by a broken line.

FIG. 4 c shows the profiles of the two currents I 1 and I 2 which flow as a result of the voltages U A1 and U A2 which are shown in FIG. 4 b ) and are applied, respectively, to the first winding W 1 and the second winding W 2 . In the graph, the current I is measured along the ordinate and the power P 0 required at any point in time to generate the compensation force is measured along the abscissa. The graph of the first current I 1 is drawn as a solid line, and the profile of the second current I 2 is drawn as a broken line. In addition, the profile of the compensation current Icp, i.e. the sum of the two currents I 1 and I 2 is represented by a dotted line.

Finally, FIG. 4 d represents the profile of the total power Ptot generated in the coil 53 , with Ptot being measured along the ordinate and the power P 0 required at any point in time to generate the compensation force being measured along the abscissa. The power Ptot is constant over the entire range and is thus independent of the applied load.

FIG. 5 shows a detail of the block diagram of FIG. 3 , simplified and meant as an example, with a power distributor PD in the form of a digital potentiometer and, downstream of the latter, two partial current sources PB 1 and PB 2 .

The two partial current sources PB 1 and PB 2 are both configured in the same way. Each partial current source includes a conventional series arrangement of two operational amplifiers with feedback loops. The first partial current source PB 1 is connected to a negative current supply Vneg, and the second partial current source PB 2 is connected to a positive current supply Vpos. Through the selection of the resistance values Rx, Ry and Rz 1 , Rz 2 , the amplification factors K 1 and K 2 and the offset voltages Uof 1 and Uof 2 can be established according to the following relationships:

K ⁢ ⁢ 1 = R y R x K ⁢ ⁢ 2 = R y R x U of ⁢ ⁢ 1 = - Rz ⁢ ⁢ 1 ( Rz ⁢ ⁢ 1 + Rx ) × Vneg U of ⁢ ⁢ 2 = - Rz ⁢ ⁢ 2 ( Rz ⁢ ⁢ 2 + Rx ) × Vpos

The two coil voltages UA 1 and UA 2 are present at the outputs of the partial current sources PB 1 and PB 2 , respectively, to be delivered to the respective windings W 1 and W 2 .

The inputs of the two partial current sources are connected to the output of the digital potentiometer. The digital potentiometer, functioning as a voltage divider, generates an output voltage which lies between the position regulator output signal UU and ground potential.

In the power distributor the two control voltages U H1 and U H2 described above are determined from the control quantity x and the previously described position regulator output signal U U according to the following equations:

U H1 =U U ×x+U of1

U H2 =U U ×(1− x )+ U of2

FIG. 6 represents a strongly simplified block diagram of a further embodiment with switching elements S 1 , S 2 , which are controlled and/or regulated, respectively, by control signals SWI 1 and SWI 2 in order to supply pulse-shaped currents to the associated coils.

The first coil and the second coil are represented by their respective resistance values, i.e. the first coil resistance R 1 and the second coil resistance R 2 . The first winding with the resistance R 1 is connected on one side to a positive supply voltage Vpos and on the other side to a first connector terminal of a first switching element S 1 . The second connector terminal of the first switching element S 1 is connected by way of a first partial current source PB 1 to a negative supply voltage Vneg. The first switching element S 1 is configured as a two-way switch which has three connector terminals, wherein the third connector terminal is connected to the common circuit potential. Consequently, the connection from the first partial current source PB 1 can be switched between the first winding with the resistance R 1 and the common circuit potential. In a first switch position, the first winding with the resistance R 1 is connected to the first partial current source PB 1 , so that a first partial current I 1 flows through the first winding with the resistance R 1 .

The circuit arrangement for the second winding with the coil resistance R 2 is analogous to the first winding with the coil resistance R 1 . Thus, the connection from the second partial current source PB 2 can be switched between the second winding represented by R 2 and the common circuit potential. In this case, when the switch S 2 is in its second position, the second winding with the coil resistance R 2 is connected to the second partial current source PB 2 , so that a second partial current I 2 flows through the second winding with the resistance value R 2 .

As in the preceding examples, the two coils are connected into the circuit in such a way that for partial currents I 1 , I 2 flowing in the same direction, the forces being generated are of opposite direction (this polarity of the configuration of the coil terminations is indicated by black dots). Furthermore, the terminations of the two coils towards the switching elements S 1 and S 2 are connected to each other through a capacitor C.

The switching elements S 1 and S 2 are individually controlled, using a control voltage UctI which is entered into an analog/digital converter A/D. The analog/digital converter generates a first pulsewidth-modulated signal PWM and a second pulsewidth-modulated signal PWM*.

The first output of the analog/digital converter A/D is connected on the one hand to the first input of a first gate G 1 and on the other hand to the first input of a first delay element D 1 , so that the first gate G 1 and the first delay element D 1 receive the first pulsewidth-modulated signal PWM. A signal representing a first delay value tp 1 is entered into the second input of the first delay element D 1 in order to control the time delay of the first delay element D 1 . The output of the first delay element D 1 is connected to the second input of the first gate G 1 . The output of the first gate G 1 now delivers the first control signal SWI 1 which serves to control the first switching element S 1

Analogous to the first output, the second output of the analog/digital converter is connected to the second switching element S 2 by way of a second delay element D 2 and a second gate G 2 in order to generate from the second pulsewidth-modulated signal PWM* a second control signal SWI 2 with a second delay value tp 2 for the control of the second switching element S 2 .

With a suitable selection of the first delay value tp 1 and the second delay value tp 2 the time profile of the first control signal SWI 1 can be changed in relation to the time profile of the second control signal SWI 2 . The time profiles of the control signals SWI 1 and SWI 2 are shown in a strongly simplified form in FIG. 7 .

The basis for the time profile is a periodic process characterized by the length of a period ts. In this process, a period is defined by a periodic switching of the switching elements S 1 and S 2 and thus by the periodic switching-over between the two partial currents in the case where the switching elements S 1 and S 2 are actuated simultaneously. With the simultaneous actuation of the switching elements S 1 and S 2 , there is no additional heat being generated, so that this operating mode reflects the state of the art. Furthermore, for this operating mode, an effective time interval t* is defined during which the first partial current I 1 flows through the first winding with the coil resistance R 1 . From the effective time interval t* and the period length ts, the weighing signal w can be determined in accordance with the ratio

w = t * ts

Thus, the weighing signal w is connected directly to the applied force.

During this effective time interval t* the flow of the second partial current I 2 is interrupted, except for a charge-balancing current through the capacitor C. At the end of the effective time interval t* the two switching elements S 1 and S 2 are actuated simultaneously under this operating mode, so that during the remaining time interval ts−t*, the second partial current I 2 flows through the second coil with the coil resistance R 2 . Finally, at the end of the period ts, the two switching elements S 1 and S 2 are switched back to their original position and thus a new period begins.

In order to generate an additional amount of heat the first control signal SWI 1 is switched with a delay relative to the end of the effective time interval t*. The length of the delay is represented by a first time interval tp 1 , i.e. the switching occurs at the time t*+tp 1 . Likewise, the second control signal SWI 2 is not being switched at the beginning or at the end of a period ts, but delayed by a second time interval tp 2 , i.e. at the time ts+tp 2 . These events repeat themselves periodically at the times 0, ts, 2ts, 3ts, etc. Accordingly, there are within each period ts a first time interval tp 1 and a second time interval tp 2 during which the control signals overlap. During these time intervals, the first partial current I 1 and the second partial current I 2 are thus flowing simultaneously through the first and second windings with the resistance values R 1 and R 2 , respectively. This has the consequence that during these time intervals, heat is generated in a force-neutral way in the windings.

Preferably, the first time interval tp 1 and the second time interval tp 2 are selected to be identical, so that there is symmetry between the first control signal SWI 1 and the second control signal SWI 2 . This symmetry is achieved through a largely identical choice for the two delay values tp 1 and tp 2 in the form of a common delay value tp.

With preference, the common delay value tp is calculated according to the following formula:

tp = ts × ( - 1 2 + 1 4 + w ⁡ ( 1 - w ) )

wherein ts represents the period length and w represents the weighing signal. Through the value w, the delay value tp thus depends on the applied force L.

FIG. 8 represents an example of an electronic circuit implementation of the block schematic of FIG. 6 , wherein the switching elements S 1 and S 2 are configured as switching transistors with associated logic gates. In this arrangement, the first switching element S 1 is formed by a pair of switching transistors arranged so that they will change simultaneously but in opposite directions between their respective open and closed states and by a pair of gates assigned to the switching transistors and arranged to complement each other, so that the connection of the first partial current source PB 1 can be switched between the first winding (symbolized by its coil resistance R 1 ) and the common circuit potential in accordance with the first control signal SWI 1 . The second switching element S 2 is configured in the same way as the first switching element S 1 , so that the connection of the second partial current source PB 2 can be switched between the second winding (symbolized by its coil resistance R 2 ) and the common circuit potential in accordance with the second control signal SWI 2 .

The first control signal SWI and the second control signal SW 2 are generated from a common control signal SWI means of a delay element D and two logic gates OR and NAND. The control signal SWI can be described in this case as that control signal which by itself would cause a simultaneous switch-over between the two partial currents I 1 and I 2 . Now, this control signal SWI is sent to the first input of the gate NAND as well as to the input of the delay element D and the second input of the gate OR. The output of the delay element D is connected to the second input of the gate NAND as well as to the first input of the gate OR, so that the common control signal SWI—retarded by the delay interval tp—is used to control the gates OR and NAND. The output signal of the gate OR represents the first control signal SWI 1 , while the output signal of the gate NAND represents the second control signal SWI 2 . The two control signals SWI 1 and SWI 2 are now used to control the switching elements S 1 and S 2 , more specifically the pairs of complementary gates, as described above.

FIG. 9 represents, in strongly simplified form, the time profiles of the control signals SWI, SWI 1 and SWI 2 in accordance with in FIG. 8 . From the control signal SWI, the first control signal SWI 1 is generated by means of the gate OR and the second control signal SWI 2 is generated by means of the gate NAND. In this arrangement, the descending flank of the first control signal SWI 1 is delayed by the time interval tp in relation to the control signal SWI. Analogously, the descending flank of the second control signal SWI 2 is also delayed by the time interval tp in relation to the control signal SWI. Thus, the situation that results is similar to FIG. 7 for the case that the first delay value tp 1 and the second delay value tp 2 are selected to be equal and thus correspond to the common delay value tp. Consequently, the common delay value tp can be determined in accordance with the equation described above.