DE102019217098A1 - Control unit for determining a dead time for power electronic switches - Google Patents

Abstract

Control device (10) for controlling a half-bridge (30) of a vehicle power module, comprising an input unit (12) for receiving an operating parameter of the half-bridge (30) and measuring the commutation time and the delay times during operation of the half-bridge (30); an extraction unit (14) for extracting, based on the obtained at least one operating parameter, an adaptive dead time of the half bridge (30) from a characteristic map in which the adaptive dead time is coupled to a commutation time of the half bridge (30) and by a minimum value to prevent cross-conduction is limited; an update unit (14) for updating the commutation time and / or the minimum value during operation of the vehicle power module; and a signal unit (16) for generating a control signal for impressing the extracted adaptive dead time in the half bridge (30).

Description

TECHNICAL AREA

The present invention relates to the field of semiconductor-based power modules. In particular, the present invention relates to a control device for determining an adaptive dead time for power electronic switches, for example a converter for an electric and / or hybrid vehicle.

TECHNICAL BACKGROUND

Power electronic switches or circuits are used nowadays for a large number of applications. Power electronics play a key role, especially in the advancing electromobility. As the number of functions integrated in the vehicle increases, so does the number and complexity of the power electronics installed there.

In many power electronic circuits such as power supplies and converters, half bridges are used, which typically consist of an upper (high-side) switch and a lower (low-side) switch. The high-side switch (HS switch) can comprise a transistor, for example a field effect transistor (FET), which switches a load to a supply voltage. The low-side switch (LS switch) can also comprise a transistor, for example an FET, which switches a load to ground (or ground).

The two switches are to be controlled inverted to each other during operation. When switching on and off, there is a time delay in both an HS and an LS switch. This is due to the fact that after a gate-source voltage (V _GS ) has been applied, the drain-source voltage (V _DS ) reacts to the latter with a delay. However, the two switches have delay times of different lengths due to the different reaction speed of the drain-source voltage to the gate-source voltage when switching on and off. In addition, the control circuits for the HS switch and LS switch also have different delay times when switching on and off. As a result, the HS and LS switches cannot easily be actuated in an inverted manner with respect to one another. Otherwise, it would lead to an overlap phase during which both switches are conductive at the same time, which in turn would result in a high cross-current flowing from one of the two switches to the other switch and resembling a short circuit that destroys the entire circuit breaker. In this context, one speaks of an undesired cross-conduction of the half-bridge.

In addition, with more modern power switches, which for example include a metal-oxide-semiconductor FET (MOSFET) or a gallium nitride (GaN) -based transistor with high electron mobility (GaN-HEMT), the delay time when switching off is usually longer than when switching on, which is why the use of a dead time is necessary.

Various solutions have been proposed in the past to attempt to remedy the above problem. For example, a so-called “dead time” (or locking time) is introduced, which is defined as the period of time in which neither the HS switch nor the LS switch is closed. This prevents short-circuiting of the supply voltage and thus a cross line.

However, the power semiconductors known from the prior art have high conduction losses when power is conducted in the reverse direction when switched off. This is due to the fact that the dead time is not selected appropriately, and in many cases too large. Power semiconductors that allow reverse conduction via their channel have a significantly increased forward voltage when switched off when they conduct current in the reverse direction. For this reason, too long a dead time leads to an increase in the conduction losses in the reverse line.

The invention is therefore based on the object of optimizing the dead time in power electronic circuits so that the conduction losses in the reverse line are reduced. The dead time is optimized in such a way that it is sufficiently short that the respective power semiconductor only carries current in the reverse direction when it is switched on.

The object is achieved by a control device, a control method, an optimization method, a computer program product and a computer-readable storage medium according to the independent claims.

The vehicle power module comprises at least one power semiconductor switchable in the reverse direction. For example, semiconductors with large band gaps (English: Wide Bandgap Semiconductors) such as gallium nitride (GaN) or silicon carbide (SiC) come into consideration. The vehicle power module may include one or more metal-oxide-semiconductor field effect transistors (MOSFETs), one or more high electron mobility transistors (HEMTs). The half bridge is formed from at least some of the semiconductors contained in the vehicle power module. The half-bridge comprises a high-side switch (HS switch) and a low-side switch (LS switch). Alternatively, the half-bridge can be a so-called multilevel half-bridge.

The input interface of the control device can be designed, for example, as an analog-digital converter or as an interface of a data line (e.g. Serial Peripheral Interface, SPI). The output current of the half-bridge is determined, for example measured or estimated, during operation of the half-bridge or the vehicle power module, preferably in real time, and is thereby obtained. A current transformer, for example, can be used as the current measuring device. The output signal of the current measuring device can additionally be processed with a Kalman filter and / or an observer in order to estimate the output current, in particular in the area of the zero crossing. This increases the accuracy of the output current obtained in this way and thus improves the control of the half bridge, since the extraction of the adaptive dead time is optimized in this way. This is particularly advantageous in applications with a high output current, as it enables a much more accurate current value to be determined in the area of the zero crossing, because the resolution of the current sensor is not sufficient for an exact determination of the current value in the area of the zero crossing.

In a preferred embodiment, the junction temperature of at least one power semiconductor and the intermediate circuit voltage of the half bridge are also determined.

The extraction unit and the update unit are preferably each part of an evaluation unit. The evaluation unit comprises, for example, a microcontroller or a field-programmable gate arrangement (FPGA), or is part of the microcontroller. The extraction unit is used to extract an adaptive dead time for the half bridge from a map. The map contains a dependency of the adaptive dead time on the output current of the half bridge. The characteristics map preferably also contains a dependency of the adaptive dead time on the junction temperature of at least one power semiconductor and / or the intermediate circuit voltage.

The adaptive dead time is coupled to a commutation time of the half bridge and limited by a minimum value to prevent cross-conduction. This means that the adaptive dead time corresponds to the commutation time or is the same as long as the commutation time is greater than the minimum value of the dead time to prevent transverse conduction.

The minimum value of the adaptive dead time corresponds to a difference ("mismatch") between the HS switch and the LS switch with regard to the delay time when switching the HS switch and LS switch on and off (or vice versa). In a switching process in which the HV switch is switched on and the LS switch is switched off, delays occur in the control circuit of the circuit breakers and the circuit breakers themselves. Firstly, the respective gate electrode takes on the specified control potential (or gate voltage) with a delay. Second, the drain-source voltage reacts to the drive potential with a delay. Other delay mechanisms (e.g. in the control logic) also contribute to the overall delay when switching the half-bridge. However, these delays are not the same when switching on the HS switch and switching off the LS switch (or vice versa). The difference between the delays (mismatch) when switching on the HV switch and switching off the LS switch (or vice versa) means that both switches are switched on at the same time, which creates a so-called cross line with in some cases considerable power losses due to the associated high short-circuit current the half bridge caused. This cross line usually leads to the destruction of the half bridge and must therefore be prevented under all circumstances.

With the implementation of a minimum dead time to prevent cross-conduction, i.e. a time interval in which neither of the two switches is switched on between switching off the HV switch and switching on the LS switch (or vice versa), which at least the above. Delay or the mismatch, the cross line can be prevented. Therefore, the minimum value (the mismatch) is used as the adaptive dead time, at least in the range of higher output currents, whereby a short circuit of the half bridge can be prevented.

In the half-bridge, depending on the direction of flow of the output current, one of the two power switches, HV switch and LS-switch, represents an active switch and the other of the two power switches represents a synchronous switch conducts and is switched off at the start of the commutation time. The synchronous switch is the power switch that takes over the output current of the half-bridge after the active switch has been switched off and thereby becomes reverse-conducting. The undesirable effects of partial hard switching on and reverse conduction when switched off only occur in the synchronous switch. Both effects are to be prevented by using the adaptive dead time and thus the efficiency of the half bridge is to be increased.

The commutation time is the time that the half-bridge needs until the voltage applied across a power path of the synchronous switch, for example the drain-source voltage, has dropped from the intermediate circuit voltage to zero. Since both power switches are usually connected in parallel to the intermediate circuit capacitor, the intermediate circuit voltage is present across the synchronous switch when the active switch is switched off. The output capacitance of the synchronous switch is then discharged by an inductance (e.g. an electric machine to be controlled such as an electric motor), which causes a current in the reverse direction via the synchronous switch. Here, the voltage prevailing across the power path of the synchronous switch decreases starting from the intermediate circuit voltage and finally drops to zero. The time interval of this discharge process is called the commutation time. However, if the synchronous switch is switched on before the discharge process has ended, the voltage across the power path of the synchronous switch suddenly drops to zero. The output capacitance of the synchronous switch is suddenly discharged, which in some cases leads to considerable power losses and additional heating of the synchronous switch. This is known as partial hard turn-on or double switching. To avoid partial hard switching on, the commutation time is used between switching off the active switch and switching on the synchronous switch. In the range of low output currents, the commutation time is usually higher than the minimum dead time to prevent cross-conduction. The commutation time is therefore used as the adaptive dead time for this range. In the area of high output currents, it can happen that the commutation time is shorter than the minimum dead time to prevent cross-conduction. In this case, it is not the commutation time but the minimum dead time to prevent cross-conduction that is used as the adaptive dead time. The adaptive dead time is thus limited to the value of the minimum dead time.

The commutation time and / or the minimum value can be measured by pre-characterization on a prototype of the vehicle power module. The commutation time and / or the minimum value of the dead time to prevent cross-conduction are preferably measured as a function of the output current of the half-bridge and preferably additionally on the junction temperature of at least one power semiconductor and the intermediate circuit voltage. For each assumed value of one of the parameters (e.g. output current), the commutation time or the minimum value of the dead time to prevent cross-conduction is measured. The other parameters (output current, junction temperature, intermediate circuit voltage) can be varied. In this way, a multi-dimensional map is obtained. The commutation time can be pre-characterized, for example, by dynamically characterizing the half-bridge with the aid of a double pulse test. An inductive load (e.g. stator of an electric machine) is switched with the half bridge and the commutation time is measured. Since the inductive load is connected in parallel to one of the two circuit breakers of the half bridge, the output capacitance of the circuit breaker and the parasitic capacitance of the inductive load are connected in parallel. For this reason, the parasitic capacitance of the inductive load can influence the measured commutation time. For this reason, the most accurate results for the commutation time are obtained if the same inductive load is used for the pre-characterization that will also be switched by the half-bridge in the later application.

The update unit is used to update the commutation time and / or the minimum value during operation of the vehicle power module. For this purpose, the map contains an “initial value” of the commutation time and the minimum value of the dead time to prevent cross-conduction, which were determined in the preliminary characterization as a function of the load current and preferably also the junction temperature and the intermediate circuit voltage. The map is continuously updated when the vehicle power module is in operation.

The commutation time can preferably be measured with the aid of the voltage across the power path of the LS switch. The commutation time is preferably measured by means of an analog-digital converter which is arranged in the control device (microcontroller, FPGA) or is used as an external component.

During operation of the vehicle power module, the parameters (output current and, preferably, additional junction temperature and intermediate circuit voltage) are determined and the associated value of the commutation time or the minimum value of the dead time to prevent cross-conduction from the characteristic map is used as an adaptive dead time based on the determined value (s) of the parameters .

Based on the adaptive dead time, a control signal is generated by means of the signal unit, the control signal serving to impress the adaptive dead time of the half-bridge. The control signal can include, for example, a first pulse width modulation signal (e.g. PWM signal) for an upper switch (HS switch) of the half bridge and a second pulse width modulation signal (e.g. PWM signal) for a lower switch (LS switch) of the half bridge. The signal unit can be part of a microcontroller.

Both the transverse line and partial hard switch-on and reverse line when the synchronous switch is switched off can be prevented particularly effectively. According to the invention, a more efficient control of the vehicle power module and thus also of the electric drive by means of the multidimensional characteristics map is therefore guaranteed. The control is also reliable, since the adaptive dead time is always selected to be at least as large as the minimum dead time to prevent cross-conduction.

Advantageous refinements and developments are specified in the subclaims.

According to one embodiment, the input unit is also designed to receive parameter values measured during operation of the vehicle power module, the evaluation unit being designed to select the adaptive dead time according to the dependency contained in the map and the parameter value obtained from the map.

The evaluation unit is preferably designed to select the minimum value of the dead time for preventing cross conduction in accordance with the dependency contained in the characteristic diagram and the parameter value obtained from the characteristic diagram. This is done, for example, in that the dependency of the adaptive dead time contained in the characteristic diagram already contains a dependency of the minimum value on the parameter or parameters (output current, junction temperature, intermediate circuit voltage). The adaptive dead time thus takes into account the operating state of the vehicle power module, so that the adaptive dead time is always selected in such a way as to avoid the cross line and the partial hard switch-on as well as the reverse line when the synchronous switch is switched off. Additional losses due to partial hard switching on and reverse conduction when the synchronous switch is switched off can thereby be prevented, which is why the efficiency of the half bridge can be increased.

According to a further embodiment, the minimum value of the dead time to prevent transverse conduction is determined by characterizing the half-bridge during operation.

It is preferably an ongoing characterization during operation of the half-bridge, in which the half-bridge is characterized at regular time intervals. The regular time interval can be, for example, one day, two days, or three days. The minimum value of the dead time to prevent cross-conduction is preferably determined by measuring the delay times in the control of the half-bridge and the switch-on and switch-off delays of the power semiconductors during operation. The sum of the individual switch-on and switch-off delays is then used to calculate the difference between the total switch-off delay and the total switch-on delay of the HS switch and the LS switch (or vice versa). This difference corresponds to the minimum dead time to prevent cross conduction. The characteristics map is updated with the values obtained in this way for the minimum value or for the delay time. The adaptive dead time is therefore based on the parameter knowledge of the half-bridge gained during operation of the vehicle power module, so that the control of the vehicle power module is more reliable. By this measure, the effect of effects such as component tolerances and aging effects of the components on the minimum dead time for preventing cross conduction can be taken into account and the minimum dead time for preventing cross conduction during operation can be corrected accordingly.

According to a further embodiment, the commutation time is determined by characterizing the half-bridge during operation.

It is preferably an ongoing characterization of the half-bridge, in which the half-bridge is characterized at regular time intervals. The regular time interval can be, for example, one day, two days, or three days. The map is updated with the commutation time values obtained in this way. The adaptive dead time is therefore based on that during operation of the Vehicle power module gained parameter knowledge of the half bridge so that the control of the vehicle power module is more efficient. This measure allows the effect of effects such as component tolerances and aging effects of the components on the commutation time to be taken into account and the commutation time to be corrected during operation.

According to a further embodiment, the characteristic diagram is a multidimensional characteristic diagram in which several dependencies of the adaptive dead time on several parameters of the half-bridge are contained.

According to a further embodiment, the adaptive dead time of the commutation time is determined after the active switch of the half bridge is switched off.

When the active switch is turned off, the voltage across the synchronous switch changes very much. This strong voltage change takes a correspondingly long time with low output currents. In order to ensure that the synchronous switch is not switched on before the voltage applied to it has reached zero, which would lead to partial hard switching on, the commutation time is therefore used as the adaptive dead time in this case.

According to a further embodiment, the adaptive dead time is set to the minimum value of the dead time to prevent cross-conduction after the synchronous switch of the half-bridge is switched off.

When the synchronous switch is turned off, the forward voltage across the synchronous switch increases only slightly. This small change in voltage takes correspondingly little time. Therefore, in this case, the minimum value can be used as the adaptive dead time in order to prevent the cross-line.

Alternatively, when the synchronous switch is switched off, it is checked how the output current is in relation to a predefined threshold and the value zero. The predefined threshold preferably comprises a positive threshold and a negative threshold which is dependent on the amplitude of the output current, the frequency of the output current and the time between two sampling points of the output current. If the output current falls within a positive half-cycle between the positive threshold and zero, and / or rises within a negative half-cycle between the negative threshold and zero, the adaptive dead time is used both before switching on and after switching off the synchronous switch the commutation time is applied. This advantageously prevents the undesired partial hard switching of the half-bridge. Only when the next current zero crossing is measured does the normal application of the adaptive dead time continue, which is why the commutation time is used before the synchronous switch is switched on and the minimum dead time is used to prevent cross-conduction after the synchronous switch is switched off.

The computer program product according to the invention is designed to be loaded into a memory of a computer and comprises software code sections with which the method steps of the method according to the invention are carried out when the computer program product is running on the computer.

A program belongs to the software of a data processing system, for example an evaluation device or a computer. Software is a collective term for programs and associated data. The complement to software is hardware. Hardware describes the mechanical and electronic alignment of a data processing system. A computer is an evaluation device.

Computer program products generally comprise a sequence of instructions which, when the program is loaded, cause the hardware to carry out a specific method that leads to a specific result. When the program in question is used on a computer, the computer program product produces the inventive technical effect described above.

The computer program product according to the invention is platform-independent. That means it can be run on any computing platform. The computer program product is preferably executed on an evaluation device according to the invention for detecting the surroundings of the vehicle.

The software code sections are written in any programming language, for example in Python, Java, JavaScript, C, C ++, C #, Matlab, LabView, Objective C. Alternatively or additionally, a hardware description language (e.g. VHDL) can be used.

The computer-readable storage medium is, for example, an electronic, magnetic, optical or magneto-optical storage medium.

• 1 eine schematische Darstellung eines Steuergeräts gemäß einer Ausführungsform;
• 2 eine schematische Darstellung einer Halbbrücke, die durch das Steuergerät unter Verwendung der adaptiven Totzeit angesteuert wird;
• 3 eine schematische Darstellung zur Veranschaulichung von Verzögerungszeiten im Ansteuerkreis der Halbbrücke;
• 4 eine schematische Darstellung zur Veranschaulichung der Anwendung der adaptiven Totzeit entsprechend der Kommutierungszeit des synchronen Schalters am Beispiel des LS-Schalters der Halbbrücke;
• 5 eine schematische Darstellung der Zuordnung aktiver/synchroner Schalter der Halbbrücke abhängig von der Flussrichtung des Ausgangsstroms und Anwendung der adaptiven Totzeit für den Strombereich außerhalb des Nulldurchgangs des Ausgangsstroms der Halbbrücke; und
• 6 eine schematische Darstellung zur Veranschaulichung der Anwendung der adaptiven Totzeit im Bereich des Nulldurchgangs des Ausgangsstroms der Halbbrücke.
• 7 Ermittlung der Verzögerungszeiten des gesamten Ansteuerkreises und der Leistungshalbleiter durch Messung der Ausgangsspannung Usw der Halbbrücke im Betrieb der Halbbrücke
• 8 Ermittlung der Kommutierungszeit durch Messung der Ausgangsspannung Usw der Halbbrücke im Betrieb der Halbbrücke

Embodiments will now be described by way of example and with reference to the accompanying drawings. Show it:

• 1 a schematic representation of a control device according to an embodiment;
• 2 a schematic representation of a half bridge that is controlled by the control unit using the adaptive dead time;
• 3 a schematic representation to illustrate delay times in the control circuit of the half bridge;
• 4th a schematic representation to illustrate the application of the adaptive dead time corresponding to the commutation time of the synchronous switch using the example of the LS switch of the half bridge;
• 5 a schematic representation of the assignment of active / synchronous switches of the half bridge depending on the flow direction of the output current and application of the adaptive dead time for the current range outside the zero crossing of the output current of the half bridge; and
• 6th a schematic representation to illustrate the application of the adaptive dead time in the area of the zero crossing of the output current of the half bridge.
• 7th Determination of the delay times of the entire control circuit and the power semiconductors by measuring the output voltage Usw of the half bridge when the half bridge is in operation
• 8th Determination of the commutation time by measuring the output voltage Usw of the half bridge when the half bridge is in operation

In the figures, the same reference symbols relate to the same or functionally similar reference parts. The relevant reference parts are identified in the individual figures.

1 shows a schematic representation of a control device 10 according to one embodiment. The control unit 10 comprises an input unit 12th , an evaluation unit 14th and a signal unit 16 . The input unit 12th serves for this purpose the output current of the half bridge and preferably additionally the junction temperature of at least one power semiconductor and the intermediate circuit voltage of the half bridge 30th which in one from the control unit 10 to be controlled vehicle power module is arranged to receive. The vehicle power module can be, for example, a drive converter of an electric vehicle and / or a hybrid vehicle or a DC / DC converter. The junction temperature refers to a junction of at least one in the half bridge 30th used power semiconductor. The intermediate circuit voltage relates to a voltage applied to an intermediate circuit capacitor of the vehicle power module. The output current and preferably also the junction temperature of at least one power semiconductor and the intermediate circuit voltage are set at the half-bridge when the vehicle power module is in operation 30th measured.

As in 2 shown schematically, comprises the half bridge 30th in the embodiment shown there as an example, designed as a half-bridge, a high-side switch (HS switch) 32 and a low-side switch (LS switch) 34 , each of which is driven by a gate 322 , 342 can be controlled. The gate control 322 , 342 is in each case with the control unit 10 connected. The HV switches are on the current output side 32 and the LS switch 34 with a current measuring device 37 for measuring the output current of the half bridge 30 as well as a load 38 , which includes, for example, an electric drive (for example an electric machine such as an electric motor). The two circuit breakers 32 , 34 are to one in 2 DC link capacitor not shown connected in parallel. An input voltage (in 2 designated as “U _DC ”, where the ground of the current path is designated as “PGND”). Basically the two circuit breakers 32 , 34 switched in such a way that at any given time only one of the two circuit breakers 32 , 34 is on and the other circuit breaker 32 , 34 is turned off. The current therefore either flows from the positive connection of the intermediate circuit capacitor via the HV switch 32 a burden 38 or the current flows, driven by the load 38 , through the LS switch 34 . Current input side of the half bridge 30th is also a DC voltmeter 36 intended for measuring the intermediate circuit voltage, which is connected to the control unit 10 connected is.

The control unit 10 also includes an evaluation unit 14th , which comprises an extraction unit and an update unit, the extraction unit is used to set an adaptive dead time for the half-bridge 30th from a map stored in a storage medium 102 is pre-stored to extract. The map includes a dependence of the adaptive dead time on the load current and preferably also on the junction temperature of at least one power semiconductor and the intermediate circuit voltage of the half-bridge 30th . The map is preferably a multidimensional map in which the adaptive dead time is dependent on several parameters, namely the output current, the intermediate circuit voltage and the junction temperature of the half-bridge 30th is registered. By measuring the parameter or parameters during operation of the vehicle power module, the value of the adaptive dead time associated with the measured parameter can be read from the characteristic diagram. In this case, the output current can be determined with increased accuracy by means of a Kalman filter in order to improve the reliability of the control of the half-bridge.

The adaptive dead time is on the one hand coupled to the commutation time and on the other hand limited by a minimum value to prevent cross-conduction. This means that the adaptive dead time corresponds to the commutation time or is the same as long as the commutation time is greater than the minimum value of the dead time to prevent transverse conduction.

The commutation time and / or the minimum value of the dead time to prevent cross-conduction can be measured by pre-characterizing a prototype of the vehicle power module. In particular, the commutation time and / or the minimum value for preventing cross-conduction is measured as a function of the output current of the half-bridge. The commutation time and / or the minimum value for preventing cross-conduction is preferably also measured as a function of the junction temperature of at least one power semiconductor and the intermediate circuit voltage of the half-bridge. For each assumed value of one of the parameters (e.g. output current), the commutation time or the minimum value to prevent cross-conduction is measured. The other parameters (junction temperature, intermediate circuit voltage) can be varied. In this way, a multi-dimensional map is obtained.

The update unit is used to update the commutation time and / or the minimum value during operation of the vehicle power module. For this purpose, the map contains an “initial value” of the commutation time and the minimum value of the dead time to prevent cross-conduction, which were determined in the preliminary characterization as a function of the load current and preferably also the junction temperature and the intermediate circuit voltage. The map is continuously updated when the vehicle power module is in operation.

The commutation time is preferably measured on the LS switch. The commutation time is preferably measured by means of an analog-digital converter which is arranged in the control device (microcontroller, FPGA) or is used as an external component.

The control unit 10 also includes a signal unit 16 for generating a control signal which is used to impress the extracted adaptive dead time in the half-bridge 30th serves. The control signal can, for example, be a first pulse width modulation signal (eg PWM signal) for the HS switch 32 the half bridge 30th and a second pulse width modulation signal (eg PWM signal) for the LS switch 34 the half bridge 30th include. The signal unit can be part of a microcontroller.

3 shows a schematic representation to illustrate delay times in the control circuit of the half bridge 30th . Specifically shows 3 a switching process in which the HS switch 32 switched on and the LS switch 34 is turned off. There are delay times in the gate control 322 , 342 the circuit breaker 32 , 34 on the delay times of the control logic 3222 , 3422 , Gate driver 3224 , 3424 and gate drive circuit 3226 , 3426 are due. It should be mentioned here that 3 only the delay times of the control circuits, but not the additional delay times of the circuit breakers 32 , 34 shows.

The delay time t _{del, on} when switching on the HS switch 32 and the delay time t _del , _off when switching off the LS switch 34 however, they are not the same. The difference (mismatch) between the delay _{times t del} , _off and t _{del, on} can lead to both switches being switched on at the same time, which causes a so-called cross line with in some cases considerable power losses due to the associated high short-circuit current in the half-bridge. This case occurs when the delay time t _{del, on} when switching on the HS switch is less than the delay time t _del , _off when switching off the LS switch (or vice versa).

With the implementation of a minimum dead time to prevent cross-conduction, ie a time interval in which between switching off the HV switch 32 and switching on the LS switch 34 If both switches are switched off for at least the difference between the above-mentioned delay times t _del , _off and t _{del, on} or the mismatch of the delay times between the HS switch and the LS switch, the cross line can be prevented. Too long a dead time would lead to the synchronous switch conducting in reverse direction when it is switched off (see 4th ), which would be detrimental to efficiency. Therefore, the minimum value of the dead time to prevent cross-conduction (the mismatch of the delay times between HV switch and LS-switch) is used as the adaptive dead time, at least in the range of higher output currents, because in this area the commutation time is less than the minimum dead time to prevent cross conduction .

4th shows a schematic representation to illustrate the selection of the adaptive dead time corresponding to the commutation time of the synchronous switch using the example of the LS switch of the half bridge 30th . The undesirable effects of partial hard switching on and reverse conduction when switched off always occur on the synchronous switch. The signal curves of apply accordingly 4th analogous also for the case when the HS switch is the synchronous switch. In this case, the voltage across the power path of the HV switch, which is the difference between the intermediate circuit voltage and the output voltage of the half-bridge, is relevant for determining the commutation time 30th is.

As already mentioned, in 4th the LS switch the synchronous switch, which is why the voltage across the power path of the LS switch is considered for the selection of the adaptive dead time in accordance with the commutation time. In the upper diagram in 4th is the gate-source voltage U _{GS of} the HV switch 32 and the LS switch 34 Plotted purely schematically over time. The HS switch is before a first point in time t ₁ 32 switched on while the circuit breaker 34 is turned off. The HS switch is activated at the first point in time t1 32 switched off. At the moment the HS switch is switched off 32 lies above the circuit breaker power path 34 the intermediate circuit voltage. After that, the output capacitance of the LS switch 34 by an inductance (such as an electric machine to be controlled such as an electric motor), which sends a current in the reverse direction via the LS switch 34 drives, discharges. This takes over the power path of the LS switch 34 applied voltage Usw starting from the intermediate circuit voltage and finally drops to zero.

At a second point in time t ₂ following the first point in time t ₁ , the LS switch is activated 34 switched off. However, if the second point in time t _{2 is selected} too early, so that the LS switch 34 is switched on before the discharge process of the output capacitance of the circuit breaker 34 is completed, the voltage increases across the power path of the circuit breaker 34 abruptly. The output capacitance of the circuit breaker 34 is discharged suddenly, which can lead to considerable power losses and additional heating of the circuit breaker 34 leads. This is known as partial hard turn-on or double switching. In order to avoid the partial hard switch-on, between switching off the HV switch 32 and switching on the LS switch 34 the adaptive dead time is applied according to the commutation time. In the range of low output currents of the half-bridge, the commutation time is usually higher than the minimum dead time to prevent cross-conduction. In the lower diagram in 4th a third time t _{3 is} shown, in which the over the power path of the LS switch 34 applied voltage Usw reaches the zero value. In order to avoid partial hard switching on (dead time is less than difference t ₃ -t ₁ ) and reverse conduction in the switched-off state (dead time is greater than difference t ₃ -t ₁ ) of the synchronous switch, the commutation time is used as the adaptive dead time.

5 shows a schematic representation to illustrate the assignment of active / synchronous switches depending on the flow direction of the output current of the half-bridge 30th . In the illustration above, the output current I _{Last flows} from a node between the HV switch that functions as the active switch 32 and the circuit breaker acting as a synchronous switch 34 in the direction of the load 38 . The control voltages U _{Akt of} the active switch and U _{Syn of} the synchronous switch are shown schematically as a function of time.

The HS switches work in the illustration below 32 as a synchronous switch and the LS switch 34 as an active switch. The output current I _load flows from the load 38 in the direction of the junction between the two circuit breakers 32 , 34 . The course of the control voltages U _Syn and U _Akt is shown schematically.

The definition of which switch is the active switch and which is the synchronous switch is determined by the flow direction of the output current of the half-bridge. The active switch is the power switch that conducts the output current of the half bridge in the forward direction and to start the Commutation time is switched off. The synchronous switch is the power switch that takes over the output current of the half-bridge after the active switch has been switched off and thereby becomes reverse-conducting.

_{Four times t 1} , t ₂ , t ₃ , t _{4 are} shown in each of the two representations. Up to a first time t ₁ is the active switch (HS switch 32 in the illustration above or LS switch 34 in the figure below) switched on while the synchronous switch (LS switch 34 in the illustration above or HS switch 32 in the illustration below) is switched off. At the first point in time t ₁ , the switch becomes active 32 switched off. At a second time t ₂ following the first time t ₁ , the synchronous switch is switched on. At a third time t ₃ following the second time t ₂ , the synchronous switch is switched off again. At a fourth time t ₄ following the third time t ₃ , the active switch is switched on again.

The time differences between t ₁ and t ₂ and between t ₃ and t ₄ each correspond to the dead time, in particular the adaptive dead time of the half bridge 30th .

When the active switch is switched off, the voltage across the synchronous switch changes very strongly, this voltage change taking a comparatively long time with small output currents. Since the commutation time for small output currents is greater than the minimum value of the dead time to prevent transverse conduction, the commutation time can be used in this case as the difference between the first time t ₁ and the second time t ₂ .

If, on the other hand, the synchronous switch is switched off, the forward voltage across the synchronous switch increases only slightly. This small voltage change takes comparatively little time. Therefore, in this case, the minimum value of the dead time for preventing transverse conduction can be used as the difference between the third point in time t _{3 and the fourth point in time t4.}

In the 5 The application of the adaptive dead time described applies to the case that the output current of the half bridge is not in the range of the zero crossing. For the range of the zero crossing of the output current, the adaptive dead time is correspondingly 6th or Tab. 1 applied.

6th shows a schematic representation to illustrate a further switching pattern of the half bridge 30th . In the diagram shown there, the output current I _{load is} plotted over time t. Four sample points are shown. The time difference between the first two sampling points and also between the last two sampling points is t _{ab in} each case. In addition, two thresholds I _{S, p} and I _{S, n} for the positive and negative half-cycle of the output current I _{load of} the half-bridge are marked in the diagram shown.

In the area of the zero crossing, the difficulty often arises that an output current cannot be sampled as often as desired and at the same time the sign of the output current can change. The definition of the active and synchronous switch can change accordingly in this area. For example, it can happen that after a current sampling with a low current (e.g. +1 A) the output current changes so much until the adaptive dead time associated with this current value is applied that the output current has already changed its sign at the time of switching ( e.g. to the negative). In order to prevent undesired partial hard switching on, after reaching a certain threshold I _{S, p} (positive threshold) between t ₃ and t ₄ (see 5 ) instead of the minimum value of the dead time, the commutation time can be used to prevent transverse conductance. This means that the commutation time is used _{both between t 1} and t ₂ and between t ₃ and t _4. The switching pattern only turns off when the next current zero crossing is measured (I _load ≤ 0 A) 5 reapplied so that between t ₁ and t ₂ the commutation time and between t ₃ and t ₄ the minimum value of the dead time is used to prevent transverse conduction.

$$I_{S,n}={\widehat{\text{I}}}_{\text{ Last}}\ast \text{sin}\left(2\pi \ast f_{Sinus}\ast t_{ab}\right)$$

Analogously to this, the commutation time is used in the negative half-cycle of the output current when the negative threshold I _{S, n} is exceeded, both between t ₁ and t ₂ and between t ₃ and t _4. The switching pattern only turns off when the next current zero crossing is reached (I _load ≥ 0 A) 5 is applied again so that the commutation _{time is applied between t 1} and t ₂ and the minimum value is applied _{between t 3} and t _4. The positive or negative threshold is calculated according to the following gig. 1 or gig. 2: $${\mathrm{I.}}_{\mathrm{S.},p}={\widehat{\text{I.}}}_{\text{load}}\ast \text{sin}\left(2\pi \ast f_{\mathrm{S.}inus}\ast t_{ab}\right)$$

$${\mathrm{I.}}_{\mathrm{S.},n}={\widehat{\text{I.}}}_{\text{load}}\ast \text{sin}\left(2\pi \ast f_{\mathrm{S.}inus}\ast t_{ab}\right)$$

Here f _{sine describes} the frequency of the sinusoidal output current I _Last and t _ab the time between two sampling points of the output current of the half bridge 30th .

The application of the adaptive dead time is preferably adapted as a function of the output current I _Last relative to the thresholds I _{S, p} , I _{S, n} and to the zero value 0 A according to the following table 1: Table 1 Condition for application dead time Time ranges of the dead time application _First I _load > I _{S, p} t1-t2: t _kom , t3-t4: t _{d, min} 2. I _{S, p} > I _load > 0 A t1-t2: t _kom , t3-t4: t _kom 3. I _load = 0 A t1-t2: t _kom , t3-t4: t _{d, min} 4. I _load > I _{S, n} t1-t2: t _kom , t3-t4: t _{d, min} 5. I _{S, n} <I _load <0 A t1-t2: t _kom , t3-t4: t _kom 6. I _load = 0 A t1-t2: t _kom , t3-t4: t _{d, min}

Starting from an in 6th The course of the output current I _{load shown} , in which I _{load is} above the positive threshold I _{S, p} (condition 1 in Tab. 1), the commutation time t _{kom is} used _{between the first point in time t 1} and the second point in time t _2. If the output current I _Load drops so that I _{Load is} between 0 A and I _{S, p} (condition 2 in Tab. 1), the commutation time t _{kom is used between t 1} and t ₂ and between t ₃ and t ₄ . If the output current I _load drops further to the value 0 A (condition 3 in Tab. 1), the same application of the adaptive dead time takes place as for condition 1. The same applies in the event that the output current I _load decreases further and at the same time increases than the negative threshold I _{S, n} (condition 4 in Tab. 1). If the output current I _load rises after reaching a minimum so that it lies between 0 A and the negative threshold I _{S, n} (condition 5 in Tab. 1), the same application of the adaptive dead time as for condition 2 takes place reaches the value 0 A again (condition 6 in Tab. 1), the same application of the adaptive dead time takes place as for conditions 1, 3 and 4.

7th shows a schematic representation for determining the delay times of the entire control circuit and the power semiconductors by measuring the output voltage Usw of the half-bridge when the half-bridge is in operation. In order to determine the total delay times, the output voltage Usw of the half bridge is preferably measured at the LS switch against ground when the half bridge is in operation. For this purpose, the time difference is measured that it takes for a change in the respective control signal (output signal of the control unit 10 ) leads to a change in the output voltage Usw of the half bridge. The measurement of the total delay time is, accordingly 7th , carried out on the active switch.

The total delay times for the control circuit of the HS switch and the HS switch, hereinafter referred to as the HS path for short, are measured when the half-bridge is in operation when the HS switch is the active switch. The total switch-on delay t _{del, on, HS of} the HS path is the time it takes from switching on the activation signal for the HS switch (output signal of the control unit 10 ) lasts until the voltage Usw across the LS switch begins to rise to the intermediate circuit voltage. The total switch-off delay t _{del, off, HS of} the HS path is the time it takes from switching off the activation signal of the HS switch (output signal of the control unit 10 ) lasts until the voltage Usw across the LS switch begins to drop from its maximum value, the value of the intermediate circuit voltage.

Analogously to this, the total delay times for the control circuit of the LS switch and the LS switch, hereinafter referred to as the LS path for short, are measured when the LS switch is the active switch. The total switch-on delay t _{del, on, LS of} the LS path is the time it takes from switching on the activation signal for the LS switch (output signal of the control unit 10 ) lasts until the voltage Usw across the LS switch begins to drop to zero. The total switch-off delay t _{del, off, LS of} the LS path is the time it takes from switching off the control signal of the LS switch (output signal of the control unit) until the voltage Usw across the LS switch begins to rise from zero.

This procedure is used in the operation of the half bridge. The measured values are used to calculate the difference between the total switch-off delay of the HV switch and the total switch-on delay of the LS switch (or vice versa), which corresponds to the minimum dead time to prevent cross-conduction. The characteristic diagram, which contains the values of the adaptive dead time, is compared with the values of the minimum dead time measured during operation to prevent cross-transmission with the aid of the update unit 14th updated.

In 8th the determination of the commutation time by measuring the output voltage Usw of the half-bridge during operation of the half-bridge is shown. During the commutation, the two output capacitances of the HS switch and the LS switch are _{reloaded by the output current I load of} the half-bridge. For this reason, the commutation time with the same output current I _{Load is} always the same, regardless of whether the synchronous switch is the HV switch or the LS switch. For this reason, it is sufficient _{to measure the commutation time t kom} on only one of the two switches. The commutation time t _kom is preferably measured with the aid of the output voltage Usw of the half bridge on the LS switch to ground. When the half-bridge is in operation, after the active switch has been switched off, the synchronous switch is not switched on until the output voltage of the half-bridge has dropped from the maximum value, the intermediate circuit voltage, to zero. The control unit measures the time that elapses between the switching off of the active switch and the output voltage Usw of the half-bridge becoming zero. This time is the commutation time t _kom . This procedure is used when the half-bridge is in operation for various output currents I _load and the characteristic diagram, which contains the values of the adaptive dead time, is updated with the values of the commutation time measured during operation. In this case, the influences of the junction temperature of at least one power semiconductor and the intermediate circuit voltage on the commutation time are preferably also taken into account.

List of reference symbols

10

Control unit

102

Storage medium

12th

Input unit

14th

Evaluation unit / extraction unit / update unit

16

Signal unit

30th

Half bridge

32

Highside switch (HS switch)

322

HS gate control

3222

HS gate control logic

3224

HS gate driver

3226

HS gate control circuit

34

Lowside switch (LS switch)

342

LS gate control

3422

LS gate control logic

3424

LS gate driver

3426

LS gate control circuit

36

DC voltmeter

37

Current measuring device

38

load

Claims (11)

Control device (10) for controlling a half-bridge (30) of a vehicle power module, comprising: - an input unit (12) for receiving at least one operating parameter of the half-bridge (30) and measuring the commutation time and the delay times during operation of the half-bridge (30); - an extraction unit (14) for extracting, based on the obtained at least one operating parameter, an adaptive dead time of the half bridge (30) from a characteristic map in which the adaptive dead time is coupled to a commutation time of the half bridge (30) and by a minimum value to prevent cross-conduction the half bridge (30) is limited; - An update unit (14) for updating the commutation time and / or the minimum value with the aid of the determined delay times during operation of the vehicle power module; and - a signal unit (16) for generating a control signal for impressing the extracted adaptive dead time in the half bridge (30). Control unit (10) Claim 1 wherein the at least one operating parameter of the half bridge (30) is measured during operation of the vehicle power module. Control device (10) according to one of the preceding claims, wherein the map is a one-dimensional or multi-dimensional map in which the adaptive dead time is contained as a function of one or more parameters of the half bridge (30). Control device (10) according to one of the preceding claims, wherein the adaptive dead time corresponds to the commutation time after the active switch of the half bridge (30) is switched off. Control device (10) according to one of the preceding claims, wherein the adaptive dead time corresponds to the minimum value for preventing a cross line after the synchronous switch of the half bridge (30) is switched off. Control unit (10) according to one of the Claims 1 to 4th , wherein the adaptive dead time corresponds to the commutation time after the synchronous switch of the half bridge (30) is switched off and an output current (I _load ) of the half bridge (30) is between zero and a predefined threshold (I _{S, p,} I _{S, n} ) . Control method for controlling a half bridge (30) of a vehicle power module, comprising: - Obtaining at least one operating parameter of the half bridge (30) and measuring the commutation time and the delay times in the operation of the half bridge (30); - Extracting, based on the obtained operating parameters, an adaptive dead time of the half bridge (30) from a map in which the adaptive dead time is coupled to a commutation time of the half bridge (30) and limited by a minimum value; - Updating the commutation time and / or the minimum value with the aid of the determined delay times in the operation of the vehicle power module; and - Generating a control signal for impressing the extracted adaptive dead time in the half bridge (30). Computer program product, comprising instructions which, when the program is executed by a computer, cause the computer to use the method according to Claim 7 to execute. Computer-readable storage medium on which the computer program product is based Claim 8 is stored. Estimation of the exact value of the output current of the half bridge (30) in the area of the zero crossing with a Kalman filter or an observer to increase the accuracy of the output current determined Carrying out a pre-characterization of the commutation time and the minimum dead time to prevent cross-conduction of the half bridge (30) in order to establish initial values for the map of the adaptive dead time

Images