DE102015221636A1 - A method of operating a metal oxide semiconductor field effect transistor - Google Patents

Abstract

Proposed is a method of operating a metal oxide semiconductor field effect transistor, MOSFET (54), wherein the MOSFET (54) is operated in a switching mode, and wherein a through an S-terminal, source, and a D-port, drain , a characterized portion of the MOSFET (54) is switched from a substantially non-conductive to a substantially conductive state (21; 23) or vice versa, and wherein a value (12a; 12b; 12c) for a G port, gate , the drive current (12) supplied to the MOSFET (54) is preset in accordance with a voltage between the D terminal and the S terminal, DS voltage (14), and wherein the drive current (12) is dependent on a time course of DS voltage (14) is switched between different values (12a, 12b, 12c). In this case, at least one of the values (12a, 12b, 12c) is predetermined by means of a control (43).

Description

State of the art

The invention relates to a method according to the preamble of claim 1, and a circuit arrangement and a computer program according to the independent claims.

A patent publication in this field is the WO 00/27032.

Metal oxide semiconductor field effect transistors (MOSFET or MOSFET) belong to the field effect transistors with insulated gate, also referred to as IGFET. Although polysilicon dominates as the gate material today, the term MOSFET has been retained. Historically, "MOSFET" is synonymous with "IGFET". Furthermore, the term MOSFET may comprise a multiplicity of concrete semiconductor structures, for example a so-called "VMOS" structure or "VMOS transistor" (English: "v-groove MOS field effect transistor"). Except in linear applications, MOSFETs are used in particular for switching applications, for example for DC-DC converters and in half-bridges or full bridges for the control of electric motors.

Disclosure of the invention

The problem underlying the invention is solved by a method according to claim 1, and by a circuit arrangement and a computer program according to the independent claims. Advantageous developments are specified in the subclaims. Features which are important for the invention can also be found in the following description and in the drawings, wherein the features, both alone and in different combinations, can be important for the invention, without being explicitly referred to again.

The invention relates to a method of operating a metal-oxide-semiconductor field-effect transistor, MOSFET, wherein the MOSFET is operated in a switching mode, and wherein a portion of the MOSFET characterized by an S-terminal, source, and a D-terminal is switched from a substantially non-conductive to a substantially conductive state, or vice versa, and wherein a value for a driving current supplied to a G terminal, gate, of the MOSFET in response to a voltage between the D terminal and the S terminal , DS-voltage ("drain-source-voltage", "VDS"), is given, and wherein the drive current is switched in dependence on a time course of the DS voltage between different values. At least one of the values is predetermined by means of a regulation. By means of the method, the MOSFET can be switched on or off in a particularly defined manner. As used herein, the term "turn on" means to drive the MOSFET to the substantially conductive state. Accordingly, the term "turn off" means to drive the MOSFET into the substantially non-conductive state.

For example, the MOSFET as "VMOS transistor" (English: "v-groove MOS-field-effect-transistor") or as a "Trench FET" or as a "Trench LDMOS" (English: "lateral double-diffused MOSFET") or by other semiconductor technologies. In particular, newer MOSFETs can switch comparatively fast.

The invention has the advantage that a switching behavior of a MOSFET can be improved. In particular, this relates to power MOSFETs. In this case, a compromise between an unwanted emission of electromagnetic waves (electromagnetic compatibility, "EMC") and resulting from the switching operation switching losses can be improved. For example, the unwanted emissions in a wide frequency range between about 100 kilohertz and about 400 megahertz can be particularly pronounced.

In the case of a steepness of the D-S voltage which is substantially the same in relation to the prior art, the switching losses can possibly be reduced by means of the method presented. The steepness is also referred to below as "edge steepness". In particular, the steepness can be set to a comparatively large extent and essentially independent of component tolerances. In addition, the present method requires comparatively little additional components and can be performed comparatively stable. In particular, predetermined limit values for the emission of unwanted electromagnetic waves can be maintained comparatively well.

A further advantage of the method according to the invention is that in each case the drive current fed into the G terminal of the MOSFET is predetermined. It is not necessary in this case to determine and / or specifically modify a voltage applied or to be applied between the G terminal and the S terminal, G-S voltage (English: "gate-source-voltage"). Instead, the D-S voltage is determined and used as an indicator of driver current switching. As a result, the method according to the invention can be carried out more simply and more accurately.

Furthermore, the control allows a comparatively small tolerance of the switching operation of the MOSFET characterizing quantities. The control can be dimensioned particularly stable. A bandwidth characterizing the regulation can be comparatively small. In addition, digital logic devices used for control do not require extremely fast switching times.

Furthermore, it is possible to arrange the control and / or logic functions for specifying the values and / or driver stages for generating the drive current in a common integrated circuit, for example in an ASIC ("application-specific integrated circuit").

In a preferred embodiment of the method, different values of the drive current are also fed in succession to the G terminal of the MOSFET for a respective switching operation. However, only one of these values, which is fed into the MOSFET during a period characterized by the so-called "Miller effect" (see below), is predetermined by means of the regulation. For other periods of the switching process, the values of the drive current are indeed given individually, but the values are fixed and thus set independently of an actual switching behavior of the MOSFETs. Preferably, these values are comparatively large, so that a "dead time" of the switching process remains low and switching losses are minimized.

In a further refinement, an edge slope of the D-S voltage determined from the time profile is compared with a desired value. The slew rate characterizes a slew rate or drop rate of the D-S voltage during a respective shift. By comparing with the setpoint, the edge steepness can be evaluated particularly easily, in particular for the control described above.

In a further refinement, the control takes place as a function of a time difference, wherein the time difference is due to leaving the substantially non-conductive state and reaching the substantially conductive state, or by leaving the substantially conductive state and reaching the substantially non-conductive state is characterized. The said leaving or reaching of a respective state can be determined, for example, by a measurement of the D-S voltage or a measurement of a current flowing via the D connection or the S connection. As a result, a particularly simple criterion for the control is proposed, which is possible without an explicit determination of the edge steepness.

In a further refinement, the drive current is switched between different values as a function of a current time characteristic of the D-S voltage. The term "current time course" here means that the time profile of the D-S voltage is determined during each switching operation of the MOSFET and is used in the same switching process as a criterion for switching the drive current. As a result, the driver current can be switched over very precisely in terms of time.

In a further embodiment of the method, a first control is provided for switching the MOSFET from the substantially non-conducting state to the essentially conducting state, and it is a second control for switching the MOSFET from the essentially conducting state to the one in the Essentially non-conductive state. As a result, a possibly different switching behavior of the MOSFET for switching on and off can be taken into account by respectively different values of the driver current, whereby the switching process is improved. In a simplified alternative, only a single control is used, for example only for turning on the MOSFET, with the same drive currents being given for turning on and turning off the MOSFET, respectively.

In one embodiment of the method, a first value for the drive current for a first state of the MOSFET is predetermined if the MOSFET is to be switched into the substantially conductive state, but substantially not yet conduct, and a second value for the driver current for predefine a second state of the MOSFET when the MOSFET is in a transient state between the substantially non-conducting and the substantially conducting state, and a third value for the driving current for a third state of the MOSFET is predetermined when the MOSFET is substantially conducting , By these method steps, an advantageous switching of the MOSFET is effected.

In the first state, the MOSFET is substantially in a non-conductive state. In this case, a voltage between the G terminal and the S terminal, GS voltage (also called "VGS") is smaller than a so-called "plateau voltage", corresponding to a so-called "Miller Plateau". The drive current to be fed into the G terminal essentially charges a gate-source capacitance of the MOSFET. The first state also characterizes a so-called "dead time" of the switching process. Although the DC voltage increases, but the DS voltage (also called "VDS") is essentially constant and corresponds for example to an operating voltage of the MOSFET.

In the second state, the G-S voltage substantially has the said plateau voltage. In the present case, this is characterized by the effect of a gate-drain capacitance or by the so-called "Miller effect". In this case, the D-S voltage already changes in the desired switching direction, however, a substantial portion of the current supplied to the G terminal driver current is temporarily required for reloading the gate-drain capacitance. The value of this drive current is specified by means of said control.

In the third state, the D-S voltage has substantially reached its final value, and accordingly, the MOSFET is substantially in the conducting state. Because the gate-drain capacitance is substantially recharged, subsequently the G-S voltage may increase beyond the plateau voltage. In this case, an on-resistance ("RDSon") defined between the D terminal and the S terminal becomes comparatively small, the electrical losses in the MOSFET correspondingly decreasing.

Furthermore, for turning on the MOSFET, it may be provided that a switching from the first to the second value for the drive current takes place when the DS voltage is smaller by a first threshold than the DS voltage in the first state, and that a Switching from the second to the third value for the drive current then occurs when the DS voltage is less than a sum of the DS voltage in the third state plus a second threshold. As a result, the switching can advantageously take place as a function of the time course of the D-S voltage, which has comparatively large changes during the switching operation and is thus comparatively simple and precisely determinable. In particular, it is not necessary to determine the G-S voltage.

By way of example, the first threshold value is dimensioned such that when the MOSFET is switched on, the DS voltage is changed so far with respect to a value in the first state that the change is greater than possible tolerances and / or interference signals, and therefore from the change the DS voltage can be closed sufficiently safely to the beginning switch-on. In a comparable way, the second threshold can be measured.

In one embodiment of the method, a first value for the drive current for a first state of the MOSFET is predetermined when the MOSFET is to be switched to the essentially nonconducting state, but essentially still conducts, and a second value for the drive current is generated for a second state of the MOSFET, when the MOSFET is in a transient state between the substantially conductive and the substantially non-conductive state, and a third value for the drive current for a third state of the MOSFET is given, when the MOSFET in Essentially does not conduct. By means of these method steps, an advantageous switching off of the MOSFET is effected.

The first, second and third states passed when the MOSFET is switched off are at least approximately comparable to the first, second and third states described above for switching on, but the order is reversed or the designations "first state" and "third state "are reversed.

Furthermore, it can be provided for switching off the MOSFET that switching from the first to the second value for the drive current takes place when the DS voltage is greater by a first threshold than the DS voltage in the first state, and that a switching from the second to the third value for the drive current then takes place when the DS voltage is greater than a difference between the DS voltage in the third state and a second threshold value. This results in comparable advantages, as described above for turning on the MOSFET.

A sizing of the first and second thresholds can be done in a similar manner to turn off the MOSFET as described above for turning on the MOSFET.

The drive current for the MOSFET is preferably generated by means of a driver stage designed as a controllable or at least an adjustable current source. The driver stage is designed in each case to feed a charging current into the G terminal (switching on of the MOSFET) or to remove a discharging current from the G terminal (switching off the MOSFET). The respective signs of the drive current may also depend on a doping or polarity of the MOSFET, apart from the direction of the switching operation. Said power source may comprise a plurality of individual power sources. The switching between the three states described above can be done by means of a logic ("sequence control"). A concrete realization of this sequential control can take place by means of discrete components and / or by means of a computer program.

Furthermore, it can be provided that a first comparison value for the DS voltage is predetermined and an associated first time is determined, and that a second comparison value for the DS voltage is specified and an associated second time is determined, and that from the first and second comparison value for the DS voltage and the associated first and second time, the edge steepness of the DS voltage is determined. This can be determined in a simple and precise manner a characterizing the switching process criterion.

In one embodiment of the method, the first comparison value corresponds to a potential generated by means of the first threshold value described above, and the second comparison value corresponds to a potential generated by means of the second threshold value described above.

Preferably, the edge steepness of the D-S voltage is a controlled variable for the control, and the second value for the drive current is a control variable for the control. As a result, a particularly stable control can be made possible. Alternatively, the slope of the D-S voltage is the control variable, and a difference between the second value for the drive current and a pilot value is the control variable. This alternative is referred to as "feedforward control" or as "feedforward control" and can further simplify and improve the control.

In one embodiment of the method, the above-described first and / or second threshold value against which the DS voltage is respectively compared in order to switch the injected driver current between different values, depending on the ascertained edge steepness of the DS voltage by means of a Regulation specified. This embodiment of the method can be carried out alternatively or in addition to the described control of the second value for the drive current.

Furthermore, it can be provided that the value for the driver current is a digital quantity and / or is determined by means of digital methods. This can be done for the first and / or the second and / or the third value for the driver current. Thus, the accuracy of the method can be further improved.

In one embodiment of the method, a time constant of the control is greater than a mean time interval between two switching operations of the MOSFET. This results in a particularly robust and stable control, wherein a dynamics of the control is comparatively small, whereby the operation is improved. In particular, the second value for the drive current given by the control is essentially constant during a single respective switching operation of the MOSFET. However, if required during operation of the MOSFET, the control may track the second value for the drive current from switching to switching (correspondingly slow). This results in a comparatively simple implementation, the control having only a comparatively small bandwidth. In addition, the method does not require extremely fast analog or digital components.

Furthermore, the invention relates to a circuit arrangement for operating at least one electromagnetic actuator, in particular for a fuel injection system for an internal combustion engine. In this case, the circuit arrangement has at least one MOSFET for switching a load, in particular a magnetic coil of the electromagnetic actuator, to an operating voltage. In addition, the circuit arrangement has means for carrying out the method according to the invention in at least one of the embodiments described above. The principle according to the invention can also be used quite generally to improve a switching operation of a MOSFET and is not necessarily limited to the switching of electromagnetic actuators or inductive loads.

In one embodiment, the circuit arrangement is used to control one or more other electrical consumers by means of a switching operation. The other electrical consumers can be any actuators, solenoids, transformers, resistive or reactive loads and the like. This is particularly advantageous for applications in motor vehicles. There are comparable advantages as described above.

Furthermore, the invention relates to a computer program which is designed to carry out the method in at least one of the embodiments described above. There are comparable advantages as described above.

Furthermore, the invention relates to a method for operating an insulated gate bipolar transistor, IGBT, (Insulated gate bipolar transistor), wherein the IGBT is operated in a switching operation, and wherein a through an E-terminal, emitter, and a C terminal, collector, characterized portion of the IGBT is switched from a substantially non-conductive to a substantially conductive state, or vice versa, and wherein a value for a driving current supplied to a G terminal, gate, of the IGBT in response to a Voltage between the C terminal and the E terminal, CE Voltage (English: "collector-emitter-voltage") is given, and wherein the drive current is switched in dependence on a time course of the CE voltage between different values. At least one of the values is predetermined by means of a regulation. By means of the method, the IGBT can be switched on or off in a particularly defined manner.

In particular, the above-described embodiments of the method for operating the MOSFET are mutatis mutandis applicable to the IGBT.

Hereinafter, exemplary embodiments of the invention will be explained with reference to the drawings. In the drawing show:

1 a diagram with a time course of a gate voltage, a gate current and a DS voltage of a MOSFET;

2 a first timing diagram with a DS voltage of the MOSFET for a power-on;

3 a second time chart with a DS voltage of the MOSFET for a power-on;

4 a schematic diagram with two comparators and two thresholds;

5 a block diagram for a control circuit for determining a second value for a drive current of the MOSFET;

6 a block diagram with a detailed structure of a block of the control loop of 5 ;

7 a timing diagram for the determined by the control circuit second value of the drive current of the MOSFET; and

8th a flowchart for a method for operating the MOSFET. The same reference numerals are used for functionally equivalent elements and sizes in all figures, even in different embodiments.

The 1 shows a principle timing diagram with a between a G-port, ("gate"), and an S-connection, (source), resulting voltage, DC voltage 10 , one in the 1 not shown metal oxide semiconductor field effect transistor, ("MOSFETs") during a turn-on of the MOSFET 54 (see also 5 ). In the present case is the MOSFET 54 dimensioned for comparatively large switching powers, for example, for controlling magnetic coils of injection valves for internal combustion engines.

Furthermore, in the time chart of 1 a driver current injected into the G terminal 12 and a voltage between a D-terminal and the S-terminal voltage, DS-voltage 14 , the MOSFET 54 shown. The sign of the respective DC voltage 10 or DS voltage 14 or the driver current 12 may be of a respective type and / or mode of operation of the MOSFET 54 be dependent and are therefore in 1 shown only as an example.

For example, the DS voltage 14 positive and thus a potential of the D-terminal is positive in relation to a ground potential.

In one in the 1 The coordinate system shown corresponds to the (horizontal) abscissa of a time t, and the (vertical) ordinate corresponds to a voltage or a current. Shown is a time zero (left in 1 ), at which the switch-on starts and a time 16 (right in 1 ), in which the turn-on is substantially completed.

The switch-on according to 1 is due to three temporally consecutive states 21 . 22 and 23 of the MOSFET 54 Characterized: A first state 21 in which the DC voltage 10 increases monotonically and the DS voltage 14 is essentially (still) constant; a second state 22 in which the DC voltage 10 is essentially constant and the DS voltage 14 becomes monotonically smaller; a third state 23 in which the DC voltage 10 (again) increases monotonically and the DS voltage 14 is essentially constant and thus a final value 28 (please refer 3 ) is approximated.

The 1 thus characterizes a method of operating the metal-oxide-semiconductor field effect transistor, MOSFET 54 (please refer 5 ), the MOSFET 54 is operated in a switching mode, and wherein a characterized by the S-terminal and the D-terminal portion of the MOSFET 54 from the substantially non-conductive state 21 in the substantially conductive state 23 or vice versa. Thereby, a value for the in the G-terminal of the MOSFETs 54 fed driver current 12 depending on the DS voltage 14 specified. In this case, according to the method, the driver current 12 depending on a time course of the DS voltage 14 between different values 12a . 12b . 12c switched, with at least one of the values 12a . 12b . 12c by means of a regulation 43 is specified, as will be explained in more detail below.

In the present case, the driver current 12 depending on a current time course of the DS voltage 14 between different values 12a . 12b . 12c switched. The term "current time course" here means that the time course of the DS voltage 14 every time the MOSFET is switched 54 is determined and the same switching process as a criterion for switching the driver current 12 is used.

As used herein, the term "turn on" means the MOSFET 54 in the substantially conductive state 23 to control. Accordingly, the term "turn off" means the MOSFET 54 in the substantially non-conductive state 21 to control.

To turn on the MOSFET 54 becomes a first value 12a for the driver current 12 for the first condition 21 of the MOSFET 54 given if the mosfet 54 in the substantially conductive state 23 should be switched, but essentially not yet, and it will be a second value 12b for the driver current 12 for the second state 22 of the MOSFET 54 given if the mosfet 54 in a transient state between the substantially non-conductive state 21 and the substantially conductive state 23 is, and it becomes a third value 12c for the driver current 12 for the third state 23 of the MOSFET 54 given if the mosfet 54 essentially directs.

The 2 shows another time chart, in the present case, a time course of the DS voltage 14 when the MOSFET is switched on 54 is shown. For the sake of clarity only, the abscissa is scaled with values between 0.2 μs (microseconds) and 1.2 μs, and the ordinate is scaled with values between 0 V (volts) and 15 V, by way of example.

Is shown in 2 a switch-on of the MOSFETs 54 as he, for example, at a substantially constant drive current 12 would result if the inventive method would not be applied. It can be seen that the switch-on process takes place comparatively quickly, in this case approximately within 0.1 μs. However, unwanted electromagnetic waves can be generated and emitted in a no longer permissible strength.

The 3 shows another timing diagram similar to the 2 but with the drive current 12 at least in the state 22 of the MOSFET 54 is specified according to the method. Compare to this in 1 the three states 21 . 22 and 23 as well as the associated values 12a . 12b and 12c the driver current 12 , Furthermore, in the 3 a first threshold 24a and a second threshold 24b drawn, which each have an initial value 26 or a final value 28 the DS voltage 14 Respectively.

According to the method it is provided that a changeover from the first value 12a to the second value 12b for the driver current 12 then takes place when a value of the DS voltage 14 around the first threshold 24a is less than a value of the DS voltage 14 in the first state 21 (compare the initial value 26 ), and that switching from the second value 12b to the third value 12c for the driver current 12 then takes place when the value of the DS voltage 14 is less than a sum of the DS voltage 14 in the third state 23 (compare the final value 28 ) plus the second threshold 24b ,

It is also in 3 to recognize that a first comparison value 14a for the DS voltage 14 predetermined and a corresponding first time t1 are determined, and that a second comparison value 14b for the DS voltage 14 predetermined and a corresponding second time t2 are determined, and that from the first and second comparison value 14a and 14b for the DS voltage 14 and the associated first and second time t1 and t2 a slope 15 the DS voltage 14 is determined. The slope 15 So it is from the time history of the DS voltage 14 determined and is present by a quotient of a difference from the first and second comparison value 14a and 14b for the DS voltage 14 (Counter) and a difference from the associated first and second time t1 and t2 (denominator) characterized.

In one of the 1 to 3 Similarly, a turn-off operation of the MOSFET can also be used 54 carried out according to the method, wherein the MOSFET 54 that is, from the essentially conductive state 23 in the substantially non-conductive state 21 is switched. These are the 1 to 3 as well as the above description apply mutatis mutandis (that is, generally in the reverse time direction and order).

The 4 shows by way of example a schematic diagram as part of a logic ("flow control") for switching between the three states 21 . 22 and 23 , The circuit diagram of 4 comprises in a left in the drawing area a first comparator 30 (Comparator) and a second comparator 32 , A non-inverting terminal of the first comparator 30 is over a block 34 with an initial value 26 comparable voltage 36 connected, for example, with a potential of an operating voltage or battery voltage. The block 34 reduces the tension 36 around the first threshold 24a ,

In a preferred embodiment, the first threshold is 24a fixed and has a value of about 100 millivolts to about 1.5 volts. The same applies to the second threshold 24b , The first threshold 24a can be different to the second threshold 24b be measured.

In a preferred embodiment, the in 4 circuit shown additionally used to the edge steepness 15 to investigate. It is therefore only a common threshold 24a and a common threshold 24b used. The in the 4 Thus, the circuit shown is also an element of the below in the 5 illustrated blocks 56 , In an alternative embodiment, the switching between the three states takes place 21 . 22 and 23 on the one hand, and the determination of the slope 15 on the other hand, with separate circuits respectively according to the block diagram of 4 , In this alternative, it is possible to separate each threshold 24a and separate thresholds 24b to use.

Furthermore, an inverting terminal of the second comparator 32 over a block 38 with a final value 28 comparable voltage 40 connected, for example, with a ground potential of the operating voltage or battery voltage. The block 38 increases the tension 40 around the second threshold 24b , Furthermore, an inverting terminal of the first comparator 30 and a non-inverting terminal of the second comparator 32 together with a potential of the D connection or with a corresponding voltage 42 ("VD") connected. The voltage 42 thus also characterizes the DS voltage 14 ,

outputs 30a and 32a (far left in the 4 ) of the comparators 30 and 32 can be indirect or immediate for switching between states 21 . 22 and 23 be used. In the first state 21 indicates the exit 30a a (digital) value "0", and the output 32a has a value of "1". In the second state 22 indicates the exit 30a a value of "1", and the output 32a also has a value of "1". In the third state 23 indicates the exit 30a a value of "1", and the output 32a has a value of "0". This information refers to the switching on of the MOSFET 54 as he is in the 1 to 3 is shown.

The 5 shows a block diagram for a control loop, wherein the second value 12b for the driver current 12 by means of a regulation 43 is determined. In an upper section of 5 are arranged consecutively from left to right: a subtractor 44 which has a setpoint 46 for the slope 15 with the current slope 15 compares; a regulator 48 which is detailed below at the 6 will be explained in more detail, and a driver stage 50 which the driver current 12 (physically) generated. This is an operation of the driver stage 50 in terms of values 12a . 12b and 12c the driver current 12 and controlled with respect to the times t1 and t2. The latter is in 5 by a control signal 52 symbolically represented.

Furthermore, in the 5 top right of the MOSFET 54 characterized by a block symbol, the MOSFET 54 the controlled system of the scheme 43 forms. On the MOSFET 54 is the DS voltage 14 available. A block 56 detects the DS voltage 14 and determines therefrom, using the times t1 and t2, the current slope 15 , which in the present case is a digital size. Besides, the tension is 36 (please refer 4 ) the block 56 supplied as input. The current slope 15 is the subtractor 44 supplied, whereby the control loop is thus closed. At an output of the subtractor 44 is accordingly a control deviation 62 (also called "E (z)") before.

Here is the slope 15 the DS voltage 14 a controlled variable 58 for the scheme 43 , where the second value 12b for the driver current 12 a manipulated variable 60 for the scheme 43 is. The second value 12b or the manipulated variable 60 are thereby by an output signal of the controller 48 characterized.

In an alternative embodiment, the slope is 15 the DS voltage 14 (also) the controlled variable 58 for the scheme 43 , where a difference value between the second value 12b for the driver current 12 and a pilot value 74 (please refer 6 ) the manipulated variable 60 for the scheme 43 is. This is referred to as "feedforward control" or as "feedforward control" and can be the control 43 improve further.

The feedforward control serves to influence possible control variables 43 to keep low. Such disturbances can be present in particular the voltage 36 (comparable to an operating voltage or battery voltage (see 4 ) and / or a current flowing through the D-terminal ("drain current"). Because of the pilot control needs the controller 48 only comparatively small by the controlled system (MOSFET 54 ) conditional deviations. A dynamic range of the controller 48 is correspondingly small, because then the voltage 36 (eg battery voltage) as pre-control value 74 can be used.

In a non-illustrated embodiment, the control is performed 43 in response to a time difference, wherein the time difference by a Leaving the substantially non-conductive state 21 and attaining the substantially conductive state 23 , or by leaving the substantially conductive state 23 and achieving the substantially non-conductive state 21 is characterized.

In one embodiment of the scheme 43 is the value 12a . 12b respectively. 12c for the driver current 12 a digital size and / or is determined by digital methods.

The regulation 43 is based on the principle that from control to control the second value 12b the driver current 12 each increased or decreased by a predetermined increment. The step size corresponds, for example, to the least significant bit ("LSB") of a digitally present second value 12b for the driver current 12 , This results in an at least similar behavior of the scheme 43 as known from classical integral controllers ("I-controllers").

6 shows the controller 48 in a slightly more detailed presentation. In an upper section of 6 are arranged consecutively from left to right: a comparator 64 which has a hysteresis at its input and produces at its output one of two possible digital values (eg +1 and -1). In the present case, the digital value +1 is assumed if the control deviation 62 is more positive (or greater) than a first reference value characterized by the hysteresis. Accordingly, the digital value -1 is assumed if the control deviation 62 is more negative (or smaller) than a second reference value characterized by the hysteresis.

In a following block 66 becomes a second value depending on the said digital value 12b the driver current 12 characterizing digital quantity increased or decreased by the least significant bit ("LSB") on a case-by-case basis. That is, if the current slope 15 less than the setpoint 46 for the slope 15 is, then becomes the second value 12b the driver current 12 increased by one LSB. Accordingly, then, if the current slope 15 greater than the setpoint 46 is, the second value 12b the driver current 12 reduced by one LSB.

This algorithm is in the 6 represented by the retraction and delay ("z-1") of a current value of said digital quantity. See a block for this 68 and a first adder 70 , A second adder 72 (in 6 far right) allows a pre-tax value 74 to the second value 12b the driver current 12 or to form the difference described above.

Because of the hysteresis of the comparator 64 a merely random change ("tilting", "fluttering") of the LSB can be prevented. Due to the stepwise change of the manipulated variable 60 the regulation is only +/- 1 LSB 43 a comparatively high stability. A preferably linear transmission behavior of the control 43 or the controller 48 Additionally prevents a possible oscillation tendency of the control loop. Accordingly, the required dynamics is comparatively low. In addition, with an (optional) use of the pre-control described above, the resulting dynamics are additionally limited to a minimum.

The 7 shows an example of a control behavior for the by the control 43 determined second value 12b the driver current 12 of the MOSFET 54 , On an abscissa of the illustrated coordinate system, a time t in milliseconds is given by way of example. A high second value 12b characterizes a comparatively high slope 15 of the MOSFET 54 at the beginning of the in the 7 shown time span.

It can be seen that in time subsequently by means of the scheme 43 the second value 12b the driver current 12 and accordingly the slope 15 as gradually becomes smaller, until the current slope 15 the setpoint 46 equivalent. In particular, it can be seen that a time constant of the control 43 is greater than a mean time interval between two switching operations of the MOSFET 54 , For example, for driving a solenoid of an electromagnetic actuator of a fuel injection valve of an internal combustion engine or for similar frequent switching operations.

A concrete realization of the scheme 43 and / or one for generating the control signal 52 (please refer 5 ) and / or to switch between the states 21 . 22 and 23 required sequence control can each be at least partially by means of discrete components ("circuitry") or an integrated circuit ("ASIC", English: "application-specific integrated circuit") and / or computer-aided by means of a computer program.

The means of 1 to 8th described regulation 43 is a so-called "predicative rule". The predicative regulation 43 determines the controlled variable (ie in particular the slope steepness 15 ) for respective switching operations of the MOSFET 54 and fits the manipulated variable 60 (ie the value 12b the driver current 12 ) for the following switching operations. It is thus a cycle-to-cycle predicative regulation 43 ,

The 8th shows a flowchart for a method for operating the MOSFET 54 , In a starting block 100 starts in the 8th presented procedure. In a following block 102 becomes the current slope 15 determined and with the setpoint 46 compared. In a following block 104 is determined by the resulting control deviation 62 the second value 12b the driver current 12 suitable (that is, with the aim of reducing the control deviation 62 ) changed. This is preferably done as described above by means of an integral controller (I controller). In further possible embodiments of the method, this is done by means of a P-controller, PI controller or PID controller, where "P" stands for "proportional" and "D" stands for "differential", based on known classical controller structures. Then the procedure branches back to the beginning of the block 102 , and so on.

In one embodiment of the method is a first scheme 43 for switching the MOSFET 54 from the substantially non-conductive state 21 in the substantially conductive state 23 provided, and it is a (similar in principle) second rule 43 for switching the MOSFET 54 from the substantially conductive state 23 in the substantially non-conductive state 21 intended. This may result in a possibly different switching behavior of the MOSFET 54 be specifically taken into account for switching on and off. However, it is also possible to have only one regulation 43 to use, for example, only for turning on the MOSFET 54 , where the same values 12a . 12b . 12c for the driver current 12 each for turning on and off the MOSFET 54 be specified.

The method can be advantageously used in the described embodiments for a circuit arrangement (not shown) for operating at least one electromagnetic actuator, in particular for a fuel injection system for an internal combustion engine. In this case, the circuit arrangement has at least one MOSFET 54 for switching a solenoid of the electromagnetic actuator to an operating voltage. In addition, the circuit arrangement has means for carrying out the method according to the invention. These means may include, for example, a computer program.

The embodiments described above for the method for operating the MOSFET 54 can also be applied according to the invention to a method for operating an insulated gate bipolar transistor, IGBT 54 , (English: "insulated gate bipolar transistor"), wherein the IGBT 54 is operated in a switching mode, and wherein a through an E-terminal, emitter (corresponding to the S-terminal, source, of the MOSFETs 54 ), and a C port, collector, (corresponding to the D port, drain, of the MOSFET 54 ) characterized section of the IGBT 54 from a substantially non-conductive to a substantially conductive state 21 . 23 or vice versa, and where a value 12a . 12b . 12c for one in a G port, gate, the IGBT 54 fed driver current 12 depending on a voltage between the C terminal and the E terminal, CE voltage 14 (English: "collector-emitter-voltage"), is given, and wherein the driver current 12 depending on a time course of the CE voltage 14 between different values 12a . 12b . 12c is switched. In doing so, at least one of the values becomes 12a . 12b . 12c by means of a regulation 43 specified.

QUOTES INCLUDE IN THE DESCRIPTION

This list of the documents listed by the applicant has been generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Cited patent literature

• WO 00/27032 [0002]

Claims (17)

Method for operating a metal oxide semiconductor field effect transistor, MOSFET ( 54 ), the MOSFET ( 54 ) is operated in a switching mode, and wherein a portion of the MOSFET (S), source, and a D, drain, characterized 54 ) from a substantially non-conductive to a substantially conductive state ( 21 ; 23 ) or vice versa, and where a value ( 12a ; 12b ; 12c ) for one in a G port, gate, of the MOSFETs ( 54 ) fed driver current ( 12 ) depending on a voltage between the D terminal and the S terminal, DS voltage ( 14 ), and wherein the drive current ( 12 ) as a function of a time course of the DS voltage ( 14 ) between different values ( 12a ; 12b ; 12c ), characterized in that at least one of the values ( 12a ; 12b ; 12c ) by means of a regulation ( 43 ) is given. Method according to claim 1, wherein an edge steepness determined from the time course ( 15 ) of the DS voltage ( 14 ) with a setpoint ( 46 ) is compared. Method according to claim 1 or 2, wherein the regulation ( 43 ) in response to a time difference, and wherein the time difference is due to leaving the substantially non-conductive state ( 21 ) and reaching the substantially conductive state ( 23 ), or by leaving the substantially conductive state ( 23 ) and achieving the substantially non-conductive state ( 21 ) is characterized. Method according to at least one of the preceding claims, wherein the drive current ( 12 ) in dependence on a current time profile of the DS voltage ( 14 ) between different values ( 12a ; 12b ; 12c ) is switched. Method according to at least one of the preceding claims, wherein a first control ( 43 ) for switching the MOSFET ( 54 ) of the substantially non-conductive state ( 21 ) in the substantially conductive state ( 23 ), and wherein a second regime ( 43 ) for switching the MOSFET ( 54 ) of the substantially conductive state ( 23 ) in the substantially non-conductive state ( 21 ) is provided. Method according to at least one of the preceding claims, wherein a first value ( 12a ) for the driver current ( 12 ) for a first state ( 21 ) of the MOSFET ( 54 ) is specified when the MOSFET ( 54 ) in the substantially conductive state ( 23 ), but essentially not yet conducting, and wherein a second value ( 12b ) for the driver current ( 12 ) for a second state ( 22 ) of the MOSFET ( 54 ) is specified when the MOSFET ( 54 ) in a transition state between the substantially non-conductive state ( 21 ) and the substantially conductive state ( 23 ), and where a third value ( 12c ) for the driver current ( 12 ) for a third state ( 23 ) of the MOSFET ( 54 ) is specified when the MOSFET ( 54 ) essentially conducts. Method according to claim 6, wherein a changeover from the first to the second value ( 12a ; 12b ) for the driver current ( 12 ) then takes place when the DS voltage ( 14 ) by a first threshold ( 24a ) is smaller than the DS voltage ( 14 ) in the first state ( 21 ), and wherein a switch from the second to the third value ( 12b ; 12c ) for the driver current ( 12 ) then takes place when the DS voltage ( 14 ) is less than a sum of the DS voltage ( 14 ) in the third state ( 23 ) plus a second threshold ( 24b ). Method according to at least one of the preceding claims, wherein a first value ( 12c ) for the driver current ( 12 ) for a first state ( 23 ) of the MOSFET ( 54 ) is specified when the MOSFET ( 54 ) in the substantially non-conductive state ( 21 ), but is still essentially conducting, and where a second value ( 12b ) for the driver current ( 12 ) for a second state ( 22 ) of the MOSFET ( 54 ) is specified when the MOSFET ( 54 ) in a transition state between the substantially conductive state ( 23 ) and the substantially non-conductive state ( 21 ), and where a third value ( 12a ) for the driver current ( 12 ) for a third state ( 21 ) of the MOSFET ( 54 ) is specified when the MOSFET ( 54 ) essentially does not conduct. Method according to claim 8, wherein a changeover from the first to the second value ( 12c ; 12b ) for the driver current ( 12 ) then takes place when the DS voltage ( 14 ) by a first threshold ( 24b ) is greater than the DS voltage ( 14 ) in the first state ( 23 ), and wherein a switch from the second to the third value ( 12b ; 12a ) for the driver current ( 12 ) then takes place when the DS voltage ( 14 ) is greater than a difference from the DS voltage ( 14 ) in the third state ( 21 ) and a second threshold ( 24a ). Method according to at least one of claims 2 to 9, wherein a first comparison value ( 14a ) for the DS voltage ( 14 ) and an associated first time (t1) are determined, and wherein a second comparison value ( 14b ) for the DS voltage ( 14 ) and an associated second time (t2) are determined, and wherein from the first and second comparison value ( 14a . 14b ) for the DS voltage ( 14 ) and the associated first and second time (t1, t2) the steepness ( 15 ) of the DS voltage ( 14 ) is determined. The method of claim 10, wherein the slope ( 15 ) of the DS voltage ( 14 ) a controlled variable ( 58 ) for the scheme ( 43 ), and wherein the second value ( 12b ) for the driver current ( 12 ) a manipulated variable ( 60 ) for the scheme ( 43 ). The method of claim 10, wherein the slope ( 15 ) of the DS voltage ( 14 ) a controlled variable ( 58 ) for the scheme ( 43 ), and wherein a difference value between the second value ( 12b ) for the driver current ( 12 ) and a pre-tax value ( 74 ) a manipulated variable ( 60 ) for the scheme ( 43 ). Method according to at least one of the preceding claims, wherein the value ( 12a ; 12b ; 12c ) for the driver current ( 12 ) is a digital size and / or determined by digital methods. Method according to at least one of the preceding claims, wherein a time constant of the regulation ( 43 ) is greater than an average time interval between two switching operations of the MOSFET ( 54 ). Method for operating an insulated gate bipolar transistor, IGBT ( 54 ), the IGBT ( 54 ), and wherein a portion of the IGBT (characterized 54 ) from a substantially non-conductive to a substantially conductive state ( 21 . 23 ) or vice versa, and where a value ( 12a . 12b . 12c ) for one in a G port, gate, of the IGBT ( 54 ) fed driver current ( 12 ) depending on a voltage between the C-terminal and the E-terminal, CE-voltage ( 14 ), and wherein the drive current ( 12 ) as a function of a time course of the CE voltage ( 14 ) between different values ( 12a . 12b . 12c ), characterized in that at least one of the values ( 12a . 12b . 12c ) by means of a regulation 43 is given. Circuit arrangement for operating at least one electromagnetic actuator, in particular for a fuel injection system for an internal combustion engine, characterized in that the circuit arrangement comprises at least one MOSFET ( 54 ) and / or at least one IGBT ( 54 ) for switching a load, in particular a magnetic coil of the electromagnetic actuator, to an operating voltage, and that the circuit arrangement comprises means for performing a method according to at least one of the preceding claims 1 to 15. Computer program, characterized in that it is designed to perform a method according to at least one of claims 1 to 15.

Images