Method for monitoring a semiconductor switch, control circuit for a semiconductor switch and switching device

Abstract

The invention relates to monitoring a potential fault in a semiconductor switch, in particular a semiconductor switch. Fault detection can be achieved by monitoring a voltage drop after the semiconductor switch has been activated with a predefined activation profile. A blanking time is provided between the start of the activation of the semiconductor switch and the start of the voltage drop monitoring; this blanking time can be adjusted depending on the activation profile of the semiconductor switch.

Description

Technical field

[0001] The present invention relates to a method for monitoring a semiconductor switch, a control circuit for a semiconductor switch and a switching device with such a control circuit. background

[0002] Semiconductor switching elements, such as MOSFETs or insulated-gate bipolar transistors (IGBTs), have been used in many areas of power electronics for some time. For example, such semiconductor switching elements are used in electrical power converters. These converters serve, for instance, to generate a single- or multi-phase alternating voltage from a direct current (DC) voltage in an electric drive system, which is then suitable for driving a connected electric machine.

[0003] In such power converters, semiconductor switching elements are typically used in so-called half-bridges, in which two semiconductor switching elements connected in series are arranged between a positive and a negative terminal of a DC voltage input. The junction between the two semiconductor switching elements can be connected to an AC voltage input for connection to the electrical machine.

[0004] During the operation of a semiconductor switching element in such a half-bridge, short-circuit detection can be implemented. This detection, based on monitoring the electrical voltage between the two power terminals of the semiconductor switch (drain-source or collector-emitter), identifies a potential fault, such as a short circuit in the semiconductor switching element. This short-circuit monitoring requires that only a fully switched-on state is monitored to avoid false detections. The switched-off state, as well as the period during the switch-on process, are ignored. This ignoring of the switch-on process is achieved through a so-called blanking time.

[0005] Furthermore, to optimize switching losses in such an arrangement, the semiconductor switching elements can be controlled depending on a selected operating point. For example, different current or voltage profiles can be provided at the control terminals of the semiconductor switching elements via suitable driver circuits.

[0006] For example, the publication DE 10 2020 202 842 A1 describes a driver circuit for a power module which provides a first gate-source voltage in a first operating mode and provides a lower second gate-source voltage at the control terminal of a power transistor in a second operating mode. Disclosure of the invention

[0007] The present invention provides a method for monitoring a semiconductor switch, a control circuit for a semiconductor switch and a switching device with the features of the independent claims.

[0008] Further advantageous embodiments are the subject of the dependent patent claims. Accordingly, the following is planned:

[0009] A method for monitoring a semiconductor switch. This method can be used, for example, to monitor a semiconductor switch in a half-bridge with two semiconductor switches connected in series. The method includes a step for controlling the semiconductor switch. This controlling step is specifically intended for closing the semiconductor switch. The method further includes a step for monitoring a voltage drop across the semiconductor switch. Here, the electrical voltage between the two power terminals, i.e., between drain and source or collector and emitter, is monitored. As an alternative to a direct voltage measurement between the two power terminals of the semiconductor switch, it is also possible to monitor a suitable electrical current, for example, the electrical current at a Kelvin source or Kelvin emitter terminal.Furthermore, the method includes a step for detecting a fault condition. Specifically, a fault condition is detected if the voltage drop across the semiconductor switch meets a predetermined fault condition. For example, this fault condition could involve the voltage drop exceeding or falling below a predefined threshold. The semiconductor switch is activated using a predefined drive profile. This predefined drive profile can be selected from a group of several predefined drive profiles. Additionally, a blanking period is provided between the start of the semiconductor switch activation and the start of monitoring the voltage drop across the semiconductor switch. This blanking period is set using the selected drive profile.In other words, the blanking time can be adjusted depending on the currently used control profile. Furthermore, the following is planned:

[0010] A control circuit for a semiconductor switch. This circuit can be used, for example, to control semiconductor switches in a half-bridge with two series-connected semiconductor switching elements. The circuit comprises a driver stage and a control unit. The driver stage is designed to provide a control signal for the semiconductor switch at a control terminal of the switch. The semiconductor switch is controlled using a predefined control profile. Specifically, the control profile can be selected from a group of several available profiles. The control unit is designed to set the predefined control profile for controlling the semiconductor switch. Furthermore, the control unit is designed to monitor the voltage drop across the semiconductor switch.As an alternative to direct voltage measurement between the two power terminals of the semiconductor switch, it is also possible to monitor a suitable electrical current, for example, the electrical current at a Kelvin source or Kelvin emitter terminal. Furthermore, the control unit is designed to detect a fault condition. In particular, the fault condition can be detected if the voltage drop across the semiconductor switch meets a predetermined fault condition. A blanking period is provided between the start of the semiconductor switch activation and the start of monitoring the voltage drop across the semiconductor switch. This blanking period is set using the selected activation profile. Finally, the following is planned:

[0011] A switching device comprising a semiconductor switching element and a control circuit according to the invention. The control circuit is designed to control the semiconductor switching element. For example, the switching device can be configured to control one or both semiconductor switching elements in a half-bridge with two semiconductor switching elements arranged in series. A separate driver stage can be provided for each semiconductor switching element of a half-bridge. The two driver stages can be controlled by a common control unit. Advantages of the invention

[0012] The present invention is based on the finding that a short-circuit monitoring system can be provided in a circuit arrangement with semiconductor switching elements, such as a half-bridge with two semiconductor switching elements arranged in series. This monitoring system is based on monitoring an electrical voltage between the two power terminals (drain-source and collector-emitter) of a semiconductor switching element. To avoid detection errors, this monitoring is only performed when the semiconductor switching element is fully switched on. The switched-off state, as well as the switch-on phase, are suppressed. In particular, the suppression during the switch-on phase is achieved by a so-called blanking time.

[0013] Furthermore, the present invention is based on the finding that different control profiles, for example current or voltage profiles, can be used to control a semiconductor switching element, in particular to switch it on, i.e., to close it. For example, the selection of a control profile can be adapted depending on the respective operating point.

[0014] If a short circuit is caused by switching on a semiconductor switching element, the point in time at which the electric current through a semiconductor switching element exceeds a critical, possibly uncontrollable value can vary depending on the control profile used.

[0015] Based on this understanding, one aspect of the present invention is to adjust the blanking time during the switch-on process of a semiconductor switching element according to the specific drive profile used for switching on the semiconductor switching element. This prevents, on the one hand, erroneous fault detection caused by an excessively short blanking time with drive profiles featuring a slow current rise. On the other hand, this concept prevents the blanking time from being extended to such an extent with a drive profile featuring a rapid current rise that, for example, in the event of a short circuit, the electric current has already exceeded a critical value by the time the blanking time has elapsed, thus potentially damaging the semiconductor switching element or preventing reliable interruption of the current.

[0016] By adjusting the blanking time interval according to the invention, reliable fault detection, for example for the detection of a short circuit, can always be achieved even with different control profiles, such as those that can be used for example with an operating point-dependent adjustment of the control profile.

[0017] According to one embodiment, an adjustable blanking time is determined from a group of several predetermined blanking times. For this purpose, the group of several predetermined blanking times can, for example, be stored in a suitable memory device. In particular, a suitable relationship between the predetermined blanking times and other parameters, such as the possible control profiles or other data related to these control profiles, can also be stored or provided in another way. In this way, a simple selection or assignment of suitable blanking times is possible.

[0018] According to one embodiment, an adjustable blanking time is determined using a signal for an operating-point-dependent configuration of the control profile. For example, a higher-level system can provide a desired operating point or other information related to a desired operating point. Using this information, a corresponding blanking time can then be selected for the respective operating point or the control profile used for that operating point. Thus, even with a variable operating point, a suitable blanking time for fault detection can always be set.

[0019] According to one embodiment, the control profiles of the group of several control profiles each comprise one or more time segments. For example, a control profile can comprise a time sequence of several current or voltage values ​​or waveforms. Accordingly, the steepness of the current rise through the semiconductor switch can also vary in each segment of this sequence. For such control profiles, an adjustable blanking time interval can be determined. The blanking time interval can be determined, in particular, using a predefined relationship between at least one time segment of the control profile and a corresponding blanking time interval. For example, it is also possible to define the blanking time interval up to a segment with a specific property, such as a specific value or waveform.For example, the blanking period can be set until the beginning or end of such a segment.

[0020] According to one embodiment, the adjustable blanking time can be determined as a function of the rate or steepness of the increase in electrical voltage between a control terminal and a power terminal of the semiconductor switch. For example, the blanking time can be started once a predetermined electrical voltage is reached at the control terminal of the semiconductor switching element. Following this variable start time, a potentially fixed blanking time can then be defined.

[0021] According to one embodiment, the method includes a step for comparing the determined adjustable blanking time with a predetermined maximum blanking time. This maximum blanking time can, for example, be fixed. In this case, the shorter of the adjustable blanking time and the predetermined maximum blanking time can be selected as the blanking time. This ensures that the blanking time never exceeds the maximum blanking time. This also guarantees that the maximum blanking time is not exceeded, even under unfavorable conditions or if the blanking time is determined incorrectly.

[0022] According to one embodiment, the method includes a step for issuing a signal if the determined adjustable blanking time exceeds the specified maximum blanking time. Such a signal can also indicate to another, external instance that the dynamic blanking time calculation has yielded a value that would exceed the maximum blanking time. Appropriate measures can then be initiated. For example, adjusting the operating point or another parameter can establish an operating state that allows or requires a shorter blanking time.

[0023] According to one embodiment, the method includes a step for applying a fixed, predetermined control profile if the determined adjustable blanking time is greater than the predetermined maximum blanking time. This allows for a very simple return to an operating state that enables safe operation with a blanking time of at most the predetermined limit.

[0024] The above embodiments and further developments can be combined with one another as appropriate. Further embodiments, further developments, and implementations of the invention also include combinations of features of the invention described previously or subsequently with regard to the exemplary embodiments, even if not explicitly mentioned. In particular, those skilled in the art will also add individual aspects as improvements or additions to the respective basic forms of the invention. Brief description of the drawings

[0025] Further features and advantages of the invention are explained below with reference to the figures. These show: Fig. 1: a schematic representation of a principle diagram of a half-bridge circuit with a control circuit according to an embodiment; Fig. 2: a schematic representation illustrating the adjustment of the blanking time period according to one embodiment; and Fig. 3: a flowchart as it underlies a method for monitoring a semiconductor switch according to one embodiment. Description of embodiments

[0026] Fig. Figure 1 shows a schematic representation of a basic circuit diagram for a circuit arrangement with a control circuit according to one embodiment. In the exemplary embodiment described below, the principle according to the invention for monitoring a semiconductor switching element and dynamically adjusting the blanking time provided is described using the example of a half-bridge with two semiconductor switching elements connected in series. However, this example is only to be understood as illustrative and expressly does not represent a limitation of the present invention. Rather, the basic principle according to the invention can also be used for other configurations of power topologies, such as multi-level inverters, high-side switches, or low-side switches.

[0027] In the embodiment shown here, a half-bridge with two semiconductor switching elements 21, 22 is arranged between a first, for example positive, connection point 31 and a second, for example negative, connection point 32. The two semiconductor switching elements 21, 22 are connected to each other at a node K. This node K can be connected to another connection point 41, for example an AC voltage connection point. For example, several such half-bridges can be used, with all half-bridges being connected to the same DC connection points 31, 32, and with the node K of each half-bridge being connected to a separate AC voltage connection point. In this way, for example, a three-phase AC voltage can be generated using three half-bridges.However, in principle any other applications are also possible, especially applications with only one half-bridge or a different number of half-bridges.

[0028] Each semiconductor switching element 21, 22 can be controlled at a control terminal by a driver circuit 11, 12. For a MOSFET, the control terminal can be, for example, the gate terminal, while the power terminals, i.e., drain and source, are connected to terminals 31 and 32, respectively, and node K. Alternatively, the control terminal of a bipolar transistor can be a base terminal, while the power terminals are the emitter and collector.

[0029] The two driver circuits 11 and 12 are controlled by a control unit 10. The control unit 10 can, for example, be a circuit arrangement with a microprocessor or microcontroller, which generates the output signals for the driver circuits 11 and 12 using input signals, such as setpoints, specifications for a current operating point, or similar. In particular, the control unit 10 can, for example, comprise an application-specific integrated circuit (ASIC).

[0030] The driver circuits 11, 12 can provide a suitable voltage or current at the control terminals for controlling, in particular for switching on, i.e., closing, the respective semiconductor switching elements 21, 22. In this context, it is also possible to vary the control of the semiconductor switching elements 21, 22 depending on the operating conditions, for example, depending on the selected operating point or other parameters. For example, the current and / or voltage at the control terminals of the semiconductor switching elements 21, 22 can be varied for control by using different resistors or resistor networks.In addition to providing a constant current or a fixed voltage, it is also possible to provide a variable current and / or voltage profile at the respective control terminal of the semiconductor switching element 21, 22 during the control process using the corresponding driver circuit 11, 12. In particular, it is also possible to specify this profile as a sequence of individual, temporally sequential segments. Furthermore, any other possibilities for varying the electrical current and / or voltage at the control terminals of the semiconductor switching elements 21, 22 are also possible. All of this is summarized below under the term "control profile".

[0031] During the activation of the semiconductor switching elements 21, 22 of a half-bridge, normally only one of the two semiconductor switching elements 21, 22 should be closed to avoid an electrical short circuit between the two terminals 31, 32 of the DC voltage connection. However, a malfunction or damage to a semiconductor switching element 21, 22 may result in the corresponding semiconductor switching element 21 or 22 remaining at least partially electrically conductive when the other semiconductor switching element 21, 22 of the half-bridge is closed. In this case, the electric current through the half-bridge increases very rapidly.

[0032] Therefore, fault or short-circuit monitoring can be provided for the half-bridge. This monitoring can be based, for example, on monitoring the electrical voltage between the two power terminals, such as drain and source or collector and emitter, after a semiconductor switching element 21, 22 is switched on or off. This monitoring is called "VDS monitoring" for MOSFETs or "desaturation monitoring" for bipolar transistors with an isolated gate terminal (IGBT). If the electrical voltage across a semiconductor switching element 21, 22 exceeds a predefined limit, this can indicate a possible fault, in particular a short circuit in the half-bridge.The corresponding semiconductor switch 21, 22 should then be reopened as quickly as possible before the electric current through the half-bridge increases to such an extent that the semiconductor switching element 21, 22 is damaged or the electric current becomes uncontrollable and opening the semiconductor switch 21, 22 is no longer possible. As an alternative to a direct voltage measurement between the two power terminals of a semiconductor switching element 21, 22, it may also be possible, for example, to monitor a voltage or current from another terminal, such as a Kelvin source or Kelvin emitter terminal of the semiconductor switching element 21, 22, and to perform fault detection based on this measurement.

[0033] However, to prevent false alarms from this monitoring, the monitoring is suspended both when the semiconductor switching element 21, 22 is switched off and during the switch-on process. The suppression of the monitoring during the switch-on process is achieved via a so-called blanking period. In conventional systems, this is usually a constant, fixed duration.

[0034] Due to the different control profiles described above for switching on and off the semiconductor switching elements 21, 22, the duty cycle can vary depending on the selected control profile. Therefore, with such a uniform blanking period, there is a risk that, with a control profile for a very short duty cycle, the electric current through the half-bridge may have already increased to such an extent in the event of a short circuit that the load on the semiconductor switching elements 21, 22 could lead to damage. In such a case, reliable opening of the still intact semiconductor switching element 21, 22 might no longer be possible. On the other hand, with a uniform blanking period that is too short for a control profile with a slow duty cycle, the switching process might not be completely finished by the end of the blanking period.This could potentially lead to a faulty short-circuit detection.

[0035] Therefore, it is intended that the blanking time be adjusted depending on the control profile used. In the simplest embodiment, for example, two different blanking times can be provided. A first, shorter blanking time can be applied to one or more control profiles with short on-times, while a second, longer blanking time is provided for one or more other control profiles with longer on-times. In principle, however, more than two different blanking times are also possible, with each blanking time being assigned one or more on-time profiles. This assignment can, for example, be stored in a memory of the control unit 10. However, any other concept is also possible.

[0036] The selection or determination of the blanking time interval can be based not only on a direct relationship between the control profile and the blanking time interval, but also on other suitable approaches. For example, the blanking time interval can also be selected or determined depending on a currently set or predefined operating point for the half-bridge.

[0037] Furthermore, it is also possible, particularly with control profiles consisting of a sequence of several consecutive sections, to determine the blanking time based on one or more sections within such a sequence. For example, the blanking time can be determined as a duration that coincides with the beginning or end of a specific section in such a sequence.

[0038] Furthermore, it is also possible, for example, to link the blanking time to the voltage conditions at the semiconductor switching elements 21, 22, in particular an electrical voltage between the control terminal and a power terminal, such as the gate-source voltage. For instance, the blanking time can be linked to exceeding a predefined threshold for the gate-source voltage. In this case, it is possible, for example, to start a further, possibly fixed, blanking time period from the variable time at which such a threshold is exceeded.

[0039] In addition to the dynamic adjustment of the blanking time described above, it is also possible to specify a maximum blanking time. In this case, the monitoring, especially the short-circuit monitoring, can be configured so that this maximum blanking time is not exceeded. In other words, the smaller of the dynamically determined blanking time and the maximum blanking time is always selected.

[0040] When using a maximum blanking time, a signal can also be output if the dynamically determined blanking time exceeds the maximum blanking time. This signal can be output, for example, as an error bit or similar. This way, an external component can also be notified that the current control operation may require a blanking time that exceeds a maximum permissible value. In response, the operating point for the half-bridge or another suitable parameter can be adjusted. This allows, for example, the operation of the half-bridge to be adapted so that subsequent monitoring based on the maximum blanking time or a shorter blanking time is possible.In particular, it is possible, for example, to automatically modify one or more operating parameters in the event that the dynamically determined blanking time span is greater than the maximum blanking time span, in order to achieve operation with reliable fault detection within the maximum blanking time span.

[0041] Fig. Figure 2 shows a schematic representation of a diagram illustrating the energy input E into a semiconductor switching element, using the example of a short circuit in a half-bridge under different drive profiles. For this purpose, a case is assumed in which one of the two semiconductor switching elements 21, 22 of a half-bridge remains closed due to a fault, while at time t0 the other semiconductor switching element 21, 22 is actively closed. Thus, at time t0, a short circuit occurs in the half-bridge, resulting in a rapidly increasing electric current through the semiconductor switching elements 21, 22. This leads to a high energy input E and a heavy load on the still intact semiconductor switching element 21, 22. The threshold E_crit represents the maximum permissible load on the semiconductor switching element 21, 22 before damage to the semiconductor switching element 21, 22 is to be expected.Therefore, the still intact semiconductor switching element 21 or 22 should be opened before this critical value is exceeded. Accordingly, the blanking time must be chosen so that a possible short circuit in the half-bridge can be detected before this critical value E_crit is exceeded.

[0042] As in Fig. As can be seen in Figure 2, with the first drive profile P1, the electrical current and thus the load E of the semiconductor switching element 21, 22 increases more than with the second drive profile P2. Therefore, a blanking period must be selected for the first drive profile P1 that is shorter than the time until time t1, at which the load exceeds the critical value E_crit. When applying the second drive profile P2, however, a longer blanking period can be selected, which only needs to end by time t2.

[0043] Fig.Figure 3 shows a flowchart illustrating a method for monitoring a semiconductor switching element 21, 22 during activation of the semiconductor switching element 21, 22 according to one embodiment. The method can, in principle, comprise any steps that may be used to implement a method analogous to the concept described above. Likewise, the circuit arrangement described above can include any components suitable for realizing the method described below.

[0044] In step S1, the semiconductor switching element 21, 22 is activated. This activation is specifically for closing the semiconductor switching element 21, 22. In step S2, a voltage drop across the semiconductor switching element is monitored. Finally, in step S3, a fault condition, in particular a short circuit, is detected. The fault condition is detected, in particular, if the voltage drop across the semiconductor switching element meets a predetermined fault condition. This fault condition can include exceeding or falling below a predefined limit value.

[0045] The semiconductor switching element 21, 22 is controlled using a predefined control profile. This control profile is selected from a group of several control profiles. As already explained, a blanking period is provided between the start of the control S1 of the semiconductor switching element 21, 22 and the start of the monitoring S2 of the voltage drop across the semiconductor switching element 21, 22. This blanking period can be set using the selected control profile.

[0046] In summary, the present invention relates to a concept for monitoring a potential fault in a semiconductor switch. Fault detection can be achieved by monitoring a voltage drop after the semiconductor switch has been activated with a predefined drive profile. A blanking time is provided between the start of the semiconductor switch activation and the start of the voltage drop monitoring; this blanking time can be adjusted depending on the drive profile of the semiconductor switch. QUOTES INCLUDED IN THE DESCRIPTION

[0000] This list of documents cited by the applicant was automatically generated and is included solely for the reader's convenience. The list is not part of the German patent or utility model application. The DPMA accepts no liability for any errors or omissions. Cited patent literature

[0000] DE 10 2020 202 842 A1

[0006]

Claims

[1] Method for monitoring a semiconductor switch (21, 22), comprising the steps: Controlling (S1) the semiconductor switch (21, 22) to close the semiconductor switch (21, 22); Monitoring (S2) of a voltage drop across the semiconductor switch (21, 22); and Detecting (S3) a fault condition if the voltage drop across the semiconductor switch (21, 22) meets a predetermined fault condition, wherein the control (S1) of the semiconductor switch is carried out using a predefined control profile selected from a group of several control profiles, wherein a blanking time interval is provided between the start of the activation (S1) of the semiconductor switch (21, 22) and the start of the monitoring (S2) of the voltage drop across the semiconductor switch (21, 22), and the blanking time period is set using the selected control profile. [2] Method according to claim 1, wherein for the blanking time period an adjustable blanking time period is determined from a group of several predetermined blanking times. [3] Method according to claim 1 or 2, wherein an adjustable blanking time interval is determined using a signaling for an operating point-dependent configuration of the control profile. [4] Method according to one of claims 1 to 3, wherein the control profiles of the group of several control profiles each comprise one or more time segments, and wherein for the blanking time period an adjustable blanking time period is determined using a predetermined relation between at least one time segment of the control profiles and a corresponding blanking time period. [5] Method according to one of claims 1 to 3, wherein for the blanking time interval an adjustable blanking time interval is determined as a function of a slope for an increase of the electrical voltage between a control terminal and a power terminal of the semiconductor switch (21, 22). [6] Method according to any one of claims 1 to 5, comprising a step for comparing the determined adjustable blanking time interval with a predetermined maximum blanking time interval; wherein the blanking time interval is in each case the smaller of the adjustable blanking time interval and the predetermined maximum blanking time interval. [7] Method according to claim 6, comprising a step to output a signal if the determined adjustable blanking time span is greater than the specified maximum blanking time span. [8] Method according to claim 6 or 7, comprising a step to apply a fixed predetermined control profile if the determined adjustable blanking time span is greater than the predetermined maximum blanking time span. [9] Control circuit for a semiconductor switch (21, 22), comprising: a driver stage (11, 12) designed to provide a control signal for the semiconductor switch (21, 22) at a control terminal of the semiconductor switch (21, 22), wherein the control of the semiconductor switch (21, 22) is carried out using a predetermined control profile selected from a group of several control profiles; a control device (10) designed to set the specified control profile for controlling the semiconductor switching element (21, 22), to monitor a voltage drop across the semiconductor switching element (21, 22) and to detect a fault condition if the voltage drop across the semiconductor switch (21, 22) meets a predetermined fault condition, wherein a blanking time period is provided between the start of the activation of the semiconductor switch (21, 22) and the start of the monitoring of the voltage drop across the semiconductor switch (21, 22), and the blanking time period is set using the selected control profile. [10] Switching device, with: a semiconductor switching element (21, 22); and a control circuit according to claim 9, which is designed to control the semiconductor switching element (21, 22).

Images