WO1998043266A1

WO1998043266A1 - Electromagnetic control device

Info

Publication number: WO1998043266A1
Application number: PCT/EP1998/001493
Authority: WO
Inventors: Karl-Günter Herrmann; Rudolf SCHÄFFER; Helmut Kunkel
Original assignee: Mannesmann Rexroth Ag
Priority date: 1997-03-21
Filing date: 1998-03-14
Publication date: 1998-10-01
Also published as: DE19711768B4; DE19711768A1

Abstract

The invention relates to an electromagnetic control device. A control circuit supplies a power element with pulse-width modulated control pulses whose pulse width repetition rate indicates control voltage intensity. The power element applies distribution voltage to an electromagnet arranged between electronic switches according to said control pulses. The control device is simplified for applications wherein the current flowing via the magnet coil should be adjusted between 0 and a maximum volume. To this end, the output pulses of a first comparator are supplied to the control input of a first electronic switch by means of a first optocoupler and the output pulses of a second comparator are supplied to the control output of a second electronic switch by means of a second optocoupler. The control voltage and a periodic voltage with a triangular course are supplied to a comparator. The other comparator is supplied with an inverted control voltage and a periodic voltage with a triangular course. The allocation of voltages supplied to the comparator inputs is adjusted in such a way that their output pulses have the same pulse width repetition rate. Output pulse overlapping time determines the duration of current flow via the electromagnet. The control device is chiefly intended for electrically controlled proportional valves. The inventive device is particularly suitable for control devices wherein the power element is supplied with power from the rectified line voltage.

Description

description

Electromagnetic actuator

The invention relates to an electromagnetic actuator with a power unit, to which pulse-width-modulated control pulses are fed from an upstream control circuit, the pulse duty factor of which is a measure of the level of a control voltage, and which applies a supply voltage to an electromagnet as a function of the control pulses, in particular for a proportional valve, according to the preamble of claim 1.

Such an actuator is known from DE 42 01 652 AI. The actuator has a power section that is designed as a bridge circuit. A transistor is arranged in each of the four bridge branches. A supply voltage is fed to one bridge diagonal and the series connection of a Hall sensor and an electromagnet is connected to the other bridge diagonal. A diode is arranged in parallel with each transistor, the direction of flow of which is opposite to the supply voltage of the power section. A control circuit is connected upstream of the power section, which converts an analog control voltage into pulse-width-modulated control pulses. The control circuit has a first comparator, to which the control voltage and a periodic voltage with a triangular shape are supplied. A second comparator of the control circuit is supplied with the control voltage, which is conducted via an inverter, and the periodic voltage with a triangular shape. Impulse voltages are present at the output of the two comparators. The pulse duty factor is a measure of the level of the control voltage. The output pulses of the two comparators are together with further pulses, which are made by inverting the output pulses of the two comparators are formed, fed to an integrated module called a signal conditioning circuit, the structure of which is not described. The manner in which the signals supplied to the signal conditioning circuit are processed is also not described in DE 42 01 652 AI. The integrated module has four outputs, the pulses of which are fed to the transistors of the power section via optocouplers. The order in which and for how long the transistors are activated is not described. So that a current driven by the supply voltage can flow through the electromagnet, two diagonally opposite transistors must conduct. Since two transistors connected in series are each directly supplied with the supply voltage of the power section, there is a risk that a current will flow through two transistors connected in series. Such a current flow would destroy these transistors and render the actuator unusable. To prevent two transistors connected in series from being conductive at the same time, special circuitry measures must therefore be taken. The known

Actuator basically allows the direction of the current flowing through the electromagnet to be reversed. In many applications, however, a reversal of the current flow is not required. It is sufficient to change the current between zero and a maximum value.

The invention has for its object to simplify an actuator of the type mentioned for applications in which the current flowing through the solenoid is to be changed between zero and a maximum value.

This object is achieved by the features characterized in claim 1. Since in each case an electronic switch and a diode in the reverse direction are connected in series, a current flow through two is supplied with the supply voltage for the power unit bridge branches connected in series not possible. This also applies if an electronic switch should always be on in the event of a fault. The actuator according to the invention only allows a current to flow from the supply voltage for the power section via the

Electromagnets when both electronic switches are conductive. If only one electronic switch is conductive and the other is blocked, the electromagnet is short-circuited via the conductive electronic switch and the diode arranged in the opposite branch of the bridge. If both electronic switches are blocked, the electromagnet is acted on with counter voltage via the two diodes. In this operating state, the current flowing through the electromagnet is broken down more quickly than when only one electronic switch is locked. The current flowing through the coil can therefore be reduced at different speeds. The duty cycle of an electronic switch cannot be reduced arbitrarily. The time-delayed activation of the two electronic switches makes it possible to significantly shorten the duration of the current flow through the electromagnet compared to the minimum on-time of the electronic switch. It is therefore possible to provide a high frequency for the pulse width modulation. By increasing the frequency, the amplitude of the vibration superimposed on the mean value of the current flowing through the electromagnet can be reduced accordingly.

An electromagnetic actuator, the power section is designed as a bridge circuit, is known from DE 39 27 972 AI. In the power section of this actuator, two transistors are different

Conductivity type arranged in diagonally opposite bridge branches. The first of the two transistors is connected to ground potential and the second to the positive connection of the supply voltage for the power section. In the other two bridge branches each have a diode whose direction of flow is opposite to the supply voltage for the power section. The first transistor is controlled by pulse-width-modulated control pulses, the frequency of which is predetermined by a periodic voltage with a triangular shape. The duty cycle of the control pulses is a measure of the level of a control voltage. It results from a comparison of the triangular voltage with the control voltage. The actual value of the current flowing through the electromagnet is compared with two limit values. The second transistor is switched into the conductive state when the current flowing through the electromagnet falls below the lower limit value and blocked when the current flowing through the electromagnet exceeds the upper limit value. During the on period of the first transistor, this results in an oscillation around an upper average current value which lies between the two limit values. The frequency and the duty cycle of this oscillation are set in accordance with the two limit values for the current and in accordance with the rate of rise and the rate of fall of the current. The frequency of this vibration is a multiple of the frequency of the triangular voltage. In order to avoid malfunctions, the resulting switch-on time of the second transistor must not be less than the minimum duration specified by the structure of the transistor. The mean value of the current flowing through the electromagnet results from the product of the upper mean current value and the pulse duty factor of the control pulses supplied to the first transistor. A fundamental frequency with a low frequency is superimposed on an oscillation with a higher frequency during the on period of the first transistor.

Advantageous developments of the invention are characterized in the subclaims. Because of the galvanic Separation of the supply voltages for the control inputs of the electronic switches from one another and from the supply voltage for the power section allows the use of two identical transistors. Transistors of the conductivity type which have the more favorable technical properties can thus be used. Warehousing is simplified by using the same transistors. The use of self-blocking transistors increases safety in the event of a fault. Galvanic isolation of the supply voltage for the control circuit from the supply voltages for the control inputs and from the supply voltage for the power section enables the control circuit to be designed independently of the level of the supply voltage for the power section. If the rectified mains voltage is used directly as the supply voltage for the power unit, no separate power supply unit - in particular no transformer - is required for the power unit. By using a current regulator, whose manipulated variable serves as a control voltage for the actuator, fluctuations in the supply voltage for the

Power section, e.g. B. due to the permissible deviations of the mains voltage from their nominal value. A smoothing element in the current value input of the current controller improves the control behavior. Exceeding a maximum permissible value of the current flowing through the electromagnet can be prevented in a simple manner by ANDing the pulse-width-modulated control pulses with an enable signal. Instead of the voltage with a triangular shape, a voltage with a sinusoidal shape can be supplied to the first two comparators. This is permissible because the duty cycle is very small for a high supply voltage for the power section. With a small duty cycle, the point of intersection between the control voltage and the sinusoidal voltage lies in the region of the zero crossings, that is to say in a region in which a sinusoidal voltage has the same time profile as a triangular voltage. If the actuator is part of a superimposed position control loop with an inductive displacement sensor, the primary winding of the displacement sensor can be supplied with a sinusoidal voltage by the same oscillator as the control circuit of the actuator. This measure saves an oscillator. In addition, there is no risk of mutual interference between two oscillators whose frequency is of the same order of magnitude.

The invention is explained in more detail below with its further details using an exemplary embodiment shown in the drawings. Show it

FIG. 1 shows a basic circuit diagram of an actuator according to the invention,

FIG. 2 shows a basic circuit diagram of the power supply unit of the actuator shown in FIG. 1,

Figure 3 shows the time course of the pulse width modulated control pulses with positive control voltage and

Figure 4 shows the time course of the pulse width modulated control pulses with a negative control voltage.

The same parts are provided with the same reference numerals.

FIG. 1 shows the basic circuit diagram of an actuator according to the invention. A power circuit 1 is preceded by a control circuit 2. The control circuit 2 converts a control voltage u _y into pulse-width-modulated control pulses for controlling the power section 1. The pulse duty factor of the pulse width modulated control pulses, ie the ratio of the switch-on time to the sum of the switch-on time and the switch-off time, is a measure of the level of the control voltage Uy. FIG. 2 shows the basic circuit diagram of a power supply unit for generating the supply voltages for the actuator. The AC line voltage is fed via lines 3 and 4 to a bridge rectifier 5. The rectified mains voltage is designated U _N. The reference potential of the voltage U _N is denoted by U _N0 and the positive potential by U _{N +} . A capacitor 6 smoothes the voltage U _N. With an AC voltage of z. B. 230 V, the rectified mains voltage is approximately 325 V. The voltage U _N serves as a supply voltage for the power unit 1. In addition, it is fed to a clocked power supply unit 7 as an input voltage. Since the circuit design of clocked power supplies is known to the person skilled in the art, the power supply 7 is only shown schematically here. An integrated component 8 converts the voltage U _N into an alternating voltage of high frequency, which is fed to the primary winding of a transformer 9 with a plurality of secondary windings which are electrically isolated from one another and from the primary winding. Rectification and smoothing form three further supply voltages U _L , U _H and U _B which are galvanically separated from one another and from the voltage U _N. The reference potential of the voltage U _L is denoted by U _L0 and the positive potential by U _{L +} . The reference potential of the voltage U _H is designated U _H0 and the positive potential U U ₊ . The reference potential of the voltage U _B is U _B0 , its positive potential is U _{B +} and its negative potential is U _B _. The voltages U _L and U _H serve as supply voltages for the control inputs of electronic switches of the power section 1. The voltage U _B serves as supply voltage for the control circuit 2. By electrically isolated feedback of one of the output voltages of the power pack 7, for. B. the positive portion of the voltage serving as supply voltage for the control circuit 2 U _B , to a dedicated input of the integrated module 8 and A voltage control can be implemented in a manner known per se in comparison with a setpoint.

As shown in FIG. 1, the power section 1 is designed as a bridge circuit with the circuit points 10 to 13. The voltage U _N is the power section 1 as

Supply voltage supplied. The node 10 is at the potential U _N0 and the node 11 is at the potential U _{N +} . The circuit points 10 and 11 form the one diagonal of the bridge circuit. A first field effect transistor 14 is arranged between the circuit points 10 and 12. A second field effect transistor 15 of the same type is arranged between the circuit points 13 and 11. The field effect transistors 14 and 15 serve as electronic switches. The connections of the field effect transistors 14 and 15 are designated in the usual way with G for GATE, D for DRAIN and S for SOURCE. The field effect transistors 14 and 15 are each controlled via the connections G and S. A diode 16 and 17 is arranged between the circuit points 10 and 13 and between the circuit points 12 and 11. The diodes 16 and 17 are arranged with respect to the voltage U _N in the reverse direction. An electromagnet 18 and a current sensor are arranged between the circuit points 12 and 13, which form the other diagonal of the bridge circuit. In the exemplary embodiment shown in FIG. 1, the current sensor is designed as a Hall sensor 19. The Hall sensor 19 converts the magnetic field caused by the current i flowing through the electromagnet 18 into a voltage u _H , which is a measure of the current flowing through the electromagnet 18. The voltage u _H is galvanically isolated from the current i. Instead of the Hall sensor 19 is also a current measurement by a current measuring resistor with subsequent electrical isolation of the voltage drop across it, for. B possible with an opto-transmitter. The voltage u _{H of} the Hall sensor 19 is the control circuit 2 as Actual value for the current i flowing through the electromagnet 18 is supplied. A level adjustment circuit 20 converts the voltage u _H into a voltage u ^ _x . In this embodiment, the level adjustment is carried out so that a voltage of 1 V corresponds to a current of 1 A.

The voltage Ui _x is fed to the actual value input of a current regulator 21 via a smoothing element 22. A setpoint voltage u ^ _{w is} supplied to the setpoint input of the current regulator 21. As with the voltage u ^ _x , the voltage u ^ _w also corresponds to a voltage of 1 V and a current of 1 A. The manipulated variable of the current regulator 21 is denoted by u _y . The voltage u _y is the control voltage for the pulse width modulation. Depending on the difference between the voltages u- _j _ _w and U _j _ _x and taking into account the time behavior of the current regulator 21, the voltage u _y assumes positive or negative values. If the time behavior of the current regulator 21 has an I component, the voltage u _y in the steady state of the current control circuit, in which the actual value is equal to the setpoint value, assumes a constant value which corresponds to the current i flowing through the electromagnet 18.

The voltage u _y is fed to an inverter 23. The output voltage of the inverter 23 is denoted by u _y _. The negative sign indicates that it is the inverted control voltage. The non-inverted control voltage is referred to below as u _{y +} . The magnitudes of the voltages u _{y +} and u _y _ are the same. The inverted control voltage u _y _ is negative if the control voltage designated u _{y +} is positive. If, on the other hand, the control voltage designated Uy ₊ is negative, the inverted control voltage u _y _ is positive. The control voltage u _{y +} is fed to the non-inverting input of a comparator 24. The inverted control voltage u _y _ is fed to the inverting input of a further comparator 25. The inverting input of the comparator 24 and the A non-inverting input of the comparator 25 is supplied with a periodic voltage u _d with a triangular shape from an oscillator 26. The voltage u-j_ _x is fed to the non-inverting input of a third comparator 27. An adjustable voltage U _{max is} supplied to the non-inverting input of the comparator 27. The tension u. _j _ _max is a measure of the maximum permissible value i _max of the current i flowing through the electromagnet 18. As with the voltage U _j _ _x , a voltage of 1 V corresponds to a current of 1 A for the voltage Ui _max . Depending on which of the two input voltages of a comparator is more positive, one of two different voltage values is present at the output of the comparator. The comparators 24, 25 and 27 are followed by trigger circuits 28, 29 and 30, respectively, which are designed as Schmitt triggers and convert the output voltages of the comparators 24, 25 and 27 into pulses which have either the value “0” or the value “1 " accept. In this exemplary embodiment, the value “0” is assigned the reference potential U _{B0 of} the voltage U _B and the value “1” is the positive potential U _{B + of} the voltage U _B. A NAND gate 31 links the output pulses of the trigger circuits 28 and 30. The output of the NAND gate 31 only assumes the value "0" if the value "1" both at the output of the trigger circuit 28 and at the output of the trigger circuit 30 pending. This is the case when the non-inverted control voltage u _{y + is} more positive than the voltage u _d and the voltage UJ_ _{X is also} less than the voltage Uj_ _max . In this case, current flows from the positive potential U _{B + of} the voltage U _B via a switch 32, which is closed during normal operation, and via the primary side of an optocoupler 33 to the reference potential U _B0 . If the output of the NAND gate 31 assumes the value "1" (corresponding to U _{B +} ), no current flows through the primary side of the optocoupler 33. If the switch 32 is open, no current can flow through the primary side of the optocoupler 33 either Current flow A current flows to the reference potential U _{L0 of} the voltage U via the primary side of the optocoupler 33 from the potential U _{L + of} the supply voltage U _L for driving the field effect transistor 14, via the secondary side of the optocoupler 33 and via the GATE SOURCE path of the field effect transistor 14 _L. This current charges the GATE-SOURCE capacitance of the field effect transistor 14 until the voltage applied to the GATE-SOURCE path is practically equal to the supply voltage U _L. The voltage present on the GATE-SOURCE path controls the field effect transistor 14 into the conductive state. Another NAND gate 34 links the output pulses of the trigger circuits 29 and 30. The output of the NAND gate 34 only assumes the value "0" if the value "1" both at the output of the trigger circuit 29 and at the output of the trigger circuit 30 " pending. This is the case when the inverted control voltage u _y _ is more positive than the voltage u _d and, moreover, the voltage u- ^ _{x is} less than the voltage Ui _max . In this case, current flows from the positive potential U _{B + of} the voltage U _B via the switch 32, which is closed during normal operation, and via the primary side of an optocoupler 35 to the reference potential U _B0 . If the output of the NAND gate 34 assumes the value "1" (corresponding to U _{B +} ), no current flows through the primary side of the optocoupler 34. If the switch 32 is open, no current can flow through the

Flow primary side of the optocoupler 35. When a current flows through the primary side of the optocoupler 35, a current flows from the potential U _{H + of} the supply voltage U _H for the control of the field effect transistor 15 via the secondary side of the optocoupler 35 and via the GATE SOURCE path of the field effect transistor 15 to the reference potential U _H0 of voltage U _H. This current charges the GATE-SOURCE capacitance of the field effect transistor 15 until the voltage across the GATE-SOURCE path is practically equal to the supply voltage U _H. The at the GATE SOURCE Line voltage applied controls the field effect transistor 15 in the conductive state. The switch 32 is controlled by an external enable signal (not shown in FIG. 1). If the enable signal is present, the switch 32 is closed and the field effect transistors 14 and 15 are driven into the conductive state as described above. In the absence of an enable signal, the switch 32 is open and the field effect transistors 14 and 15 are constantly blocked.

As FIG. 2 shows, the voltages U _N , U _L and U _{H are} electrically isolated from one another. As shown in FIG. 1, this measure allows the connection S of the field-effect transistor 14 to be connected to the reference potential U _{0 of} the voltage U _L and the connection S of the field-effect transistor 15 to be connected to the reference potential U _{H0 of} the voltage U _H. The level of the voltages U _N , U _L , U _H and U _B can be selected independently of one another. This makes it possible to optimally adapt these voltages to the respective consumers, namely the power unit 1, the control inputs of the field effect transistors 14 and 15 and the control circuit 2, and to the power requirements of these consumers.

Figures 3 and 4 consist of five diagrams arranged at the same time one below the other, which are designated by the letters a to e. Diagram a shows the time course of the non-inverted control voltage u _{y +} , the inverted control voltage u _y _ and the triangular voltage u _d as well as a sinusoidal voltage u _s which has the same slope as the voltage u _d in the region of the zero crossings. Diagram b shows the periods in which the field effect transistor 14 is conductive, and diagram c shows the periods in which the field effect transistor 15 is conductive. Diagram d shows the periods in which both field effect transistor 14 and field effect transistor 15 are conductive. The diagram e shows the periods in which neither the field effect transistor 14 nor the field effect transistor 15 is conductive. The corresponding periods are highlighted by hatching.

FIG. 3 shows the operating state in which the non-inverted control voltage u _{y + is} positive and the inverted control voltage u _y _ is negative. As long as the voltage u _{y + is} more positive than the voltage u _d (see FIG. 3a), the field effect transistor 14 is conductive (see FIG. 3b). As long as the voltage u _{d is} more positive than the voltage u _y _ (cf. FIG. 3a), the field effect transistor 15 is conductive

(see FIG. 3c). The pulse sequences shown in Figures 3b and 3c each have the same duty cycle, but are out of phase with each other. The size of the phase shift is determined by the level of the control voltage u _y and its sign. From time t ₃₀ to time t ₃₁ , the diagrams 3b and 3c overlap, ie in this period the two field effect transistors 14 and 15 are conductive. During this period, an increasing current i driven by the voltage U _N flows via the magnet coil 18. At time t _31, the field effect transistor 14 blocks. The current i flowing via the magnet coil 18 now flows further via the diode 11 and the still conducting field effect transistor 15. The current i decays slowly until the field effect transistor 14 is switched back to the conductive state at time t ₃₂ . From time t ₃₂ to time t ₃₃ , the diagrams 3b and 3c overlap again, both field effect transistors 14 and 15 are conductive (see FIG. 3d). The current i is again driven by the voltage U _N and increases until the time t ₃₃ . From this point in time, the field effect transistor 15 blocks. In this switching state, the current i flowing through the magnetic coil 18 flows via the still conductive field effect transistor 14 and the diode 16. The current i slowly decays until the field effect transistor 15 returns to the conductive one at time t ₄ Status is switched. From time t ₃₄ , the processes described above are repeated from time t ₃₀ to time t ₃₄ . The period t ₃₄ - t ₃₀ is the period of the triangular voltage u _d and thus the reciprocal of the frequency of this voltage. The current i flowing through the magnetic coil 18 is composed of a direct current component and an alternating current component superimposed thereon. The period t ₃₂ - t ₃₀ is the period of the AC component. The frequency of the AC component, ie the reciprocal of the period of the AC component, is twice the frequency of the triangular voltage u _d . The duty cycle of the AC component results from the ratio of the switch-on time t ₃₁ - t ₃₀ to the period t ₃₂ - t ₃₀ . As FIG. 3 shows, the pulse duty factor changes proportionally with the control voltage u _y . With small values of the control voltage u _y the duty cycle approaches a value of 50%. When the duty cycle of the AC component has become zero, the duty cycle of the current flowing through one of the two field effect transistors 14 and 15 is 50%. The duration of the flow phase of the current i flowing through the magnetic coil 18 can thus be chosen to be significantly shorter than the minimum on-time of an individual field effect transistor. The amplitude of the AC component of the current i is determined by the frequency of the triangular voltage u _d . The higher this frequency is selected, the smaller the amplitude of the AC component of the current i.

FIG. 4 shows the operating state in which the non-inverted control voltage u _{y + is} negative and the inverted voltage u _y _ is positive. As long as the voltage u _{y + is} more positive than the voltage u _d (cf. FIG. 4a), the field effect transistor 14 is conductive (cf. FIG. 4b). As long as the voltage u _{d is} more positive than the voltage u _y _ (cf. FIG. 4a), the field effect transistor 15 is conductive (cf. Figure 4c). The pulse sequences shown in FIGS. 4b and 4c each have the same duty cycle, but are out of phase with one another. The size of the phase shift is determined by the level of the control voltage and its sign. At time t _{40, the} voltage u _{y +} is more positive than the voltage u _d . Until then, the field effect transistor 14 is conductive (see FIG. 4b). From time t ₄₁ to time t ₄₂ , the voltage u _{d is} more positive than the voltage u _y _. The field effect transistor 15 conducts in this period. From time t ₄₃ to time t ₄ , the field effect transistor 14 conducts again. In the periods t ₀ to t ₄₁ and t ₄₂ to t ₃ (cf. FIG. 4e) neither the field effect transistor 14 nor the conduction Field effect transistor 15. In these periods, the solenoid 18 is acted upon by the diodes 16 and 17 with the voltage U _N as a counter voltage. The current i flowing through the magnetic coil 18 therefore drops more rapidly in these time periods than if one of the field effect transistors 14 and 15 conducts. The current i drops rapidly from the time t ₄₀ to the time t ₄₁ and slowly decreases from the time t ₄₁ to the time t ₄₂ . From time t ₄₂ to time t ₄₃ the current i drops again rapidly and from time t ₃ to time t ₄ again slowly. The pulse duty factor, that is the ratio of the period t ₄₁ - t ₄₀ to the period t ₄ - t ₄₀ , determines the resulting sinking speed of the current i. The possibility of reversing the sign of the control voltage u _y is particularly advantageous if the control voltage u _y - as in the exemplary embodiment shown in FIG. 1 - is the manipulated variable of a current regulator 21.

As an example for the supply voltage U _N for the

Power section 1 was based on a DC voltage of 325 V above. The following dimensioning example is based on a DC voltage of 300 V as voltage U _N to simplify the calculation. The ohmic resistance of the Solenoid 18 is 2 Ω and the largest current that is allowed to flow continuously through solenoid 18 is 3 A. This means that the solenoid is designed for a nominal voltage of 6 V. The voltage U _N applied to the magnet coil via the field effect transistors 14 and 15 is 50 times greater than the nominal voltage of the magnet coil. This means that the solenoid coil may only be briefly subjected to the voltage U _N , so that the solenoid coil 18 is not thermally destroyed. A current of 150 A is calculated from the voltage of 300 V and the resistance of 2 Ω. With pulse width modulation, the above assumptions result in an average current of 3 A if the pulse duty factor is 2%. A duty cycle of 2% is obtained e.g. B. with a duty cycle of 1 μs and a period of 50 μs. These times arise when the frequency of the voltage u _{d is} 10 kHz. The

In this case, the frequency of the AC component of the current i flowing through the magnetic coil 18 is 20 kHz. This results in a period of 50 μs. For a constant change in the current i flowing through the magnetic coil 18 from 0 to 3 A, a constant change in the duty cycle of 0 to 1 μs with a period of 50 μs is required. These considerations show that values for the control voltage u _{y are shown} in FIGS. 3 and 4 larger than they will occur in practice. Since the times t ₃₀ and t _{31 are} very close to the zero crossing of the voltage u _d , it is possible to use another voltage instead of the triangular voltage u _d , which behaves like a triangular voltage in the area of the zero crossings. As shown in FIGS. 3 and 4, the triangular voltage u _{d can be} replaced by a sinusoidal voltage u _s , which has the same time profile as the triangular voltage u _d in the region of the zero crossings. If the actuator is part of a superimposed position control loop with an inductive displacement sensor, the primary winding of the displacement sensor can be from the same oscillator as the Control circuit of the actuator can be supplied with a sinusoidal voltage. This measure saves an oscillator. In addition, the risk of ^'the interaction eliminates two oscillators whose oscillation frequency is of the same order of magnitude.

Claims

claims

1.Electromagnetic actuator with a power unit, to which pulse-width-modulated control pulses are fed from an upstream control circuit, the pulse duty factor of which is a measure of the level of a control voltage, and which applies a supply voltage to an electromagnet as a function of the control pulses, in particular for a proportional valve,

- whose control circuit has a first comparator, to which the control voltage and a periodic voltage with a triangular shape are fed, and a second comparator, to which the inverted control voltage and the periodic voltage with a triangular shape are fed, - whose power section as a bridge circuit provided with electronic switches is formed, one diagonal of which is supplied with the supply voltage for the power unit and the other diagonal of which the electromagnet is connected, characterized in that

- That two electronic switches (14, 15) are arranged in diagonally opposite bridge branches of the power section (1),

- That in the other two bridge branches of the power section (1) a diode (16, 17) is arranged, the flow direction of the supply voltage (U _N ) for the power section (1) is opposite,

- That the output pulses of the first comparator (24) to the control input of the first electronic switch (14) via a first optocoupler (33) are supplied as control pulses, - That the output pulses of the second comparator (25) are fed to the control input of a second electronic switch (15) via a second optocoupler (35) as control pulses and - that the assignment of the comparators (24, 25) supplied voltages (u _{y +} , u _y _, u _d ) for the inputs of the comparators (24, 25) is selected such that the output pulses of the comparators (24, 25) each have the same duty cycle, the time of the overlap of the output pulses being the duration of the current flow over the Electromagnet (18) determined.

2. Actuator according to claim 1, characterized in that the supply voltages (U _L , U _H ) for the control inputs of the electronic switches (14, 15) from each other and from the supply voltage (U _N ) for the power section (1) are electrically isolated.

3. Actuator according to claim 2, characterized in that the electronic switches (14, 15) are normally off field-effect transistors of the same conductivity type.

4. Actuator according to claim 2 or claim 3, characterized in that the supply voltage (U _B ) for the control circuit (2) galvanically from the supply voltages (U _H , U _L ) for the control inputs of the electronic switches (14 and 15th ) and is separated from the supply voltage (U _N ) for the power section (1).

5. Actuator according to one of claims 2 to 4, characterized in that the rectified voltage of the AC network is supplied as the supply voltage (U _N ) to the power section (1).

6. Actuator according to one of the preceding claims, characterized in that a current controller (21) has a first electrical variable (u-j_ _x ) which is a measure of the actual value of the current (i) flowing via the electromagnet (18), and a second electrical variable (UJ_ _W ), which is a measure of the desired value of the current (i) flowing via the electromagnet (18), is supplied so that the first electrical variable (UJ_ _X ) is galvanically separated from that via the electromagnet (18) flowing current (i) is separated and that the manipulated variable of the current regulator (21) is the control voltage (u _y ).

7. Actuator according to claim 6, characterized in that the first electrical variable (UJ_ _X ) is supplied to the current controller (21) via a smoothing element (22).

8. Actuator according to claim 6 or claim 7, characterized in that a third comparator (27) is provided, one input of which is the first electrical variable (u-j_ _x ) and the other input of which is a third electrical variable (u- _max ) which is a measure of the maximum permissible value (i _m aχ) of the current (i) flowing via the electromagnet (18), and that a link between the output signal of the third comparator (27) and the output pulses of the first two comparators (24 , 25) in the sense that the output pulses of the first two comparators (24, 25) are only passed on to the optocouplers (33 or 35) connected downstream if the current (i) flowing via the electromagnet (18) is smaller than the maximum allowable value (i _m aχ).

9. Actuator according to one of claims 5 to 8, characterized in that the two comparators (24, 25) instead of the periodic voltage (u _d ) with a triangular shape, a periodic voltage (u _s ) is fed with a sinusoidal shape.