Method for operating a circuit breaker, and circuit breaker

Abstract

The invention relates to a method for operating a circuit breaker (14) comprising a switching path (16), which has a semiconductor switch (22) and to which a current sensor (24) is assigned. In this case, the semiconductor switch (22) is controlled in such a way that it is operated in linear-mode operation. A power loss occurring in the semiconductor switch (22) during the linear-mode operation and a junction temperature resulting therefrom are determined, and the determined junction temperature is compared with a limit value. Control of the semiconductor switch (22) is changed depending on the comparison. The power loss is determined on the basis of the electric current conducted by means of the semiconductor switch (22) and the electric voltage dropped across the semiconductor switch (22) and on the basis of a factor which is dependent on the duration of the linear-mode operation and / or the electric voltage dropped across the semiconductor switch (22). The invention furthermore relates to a circuit breaker (14).

Description

[0001] Description

[0002] Method for operating a circuit breaker and circuit breaker

[0003] The invention relates to a method for operating a circuit breaker and to a circuit breaker. The circuit breaker comprises a switching path that has a semiconductor switch and is associated with a current sensor.

[0004] Power distribution boards are typically used to supply power to various electrical consumers. These boards are electrically connected to a main line, which carries, for example, multi-phase alternating current. The power distribution board then electrically connects other circuits assigned to the electrical consumers to the main line. To protect the electrical consumers, circuit breakers are usually provided in the power distribution system, with each circuit being assigned a corresponding circuit breaker. If a fault occurs in one of the circuits, the respective circuit breaker interrupts or limits the flow of current, thus preventing damage to the consumer and / or the (electrical) line leading to it. Such circuit breakers are also used independently of the power distribution system, for example, and are used to protect a device or line.

[0005] In the event of a fault, a distinction is usually made between an overcurrent and a short-circuit current. In the case of a short-circuit current, there is usually a low-resistance connection between the two terminals of the respective circuit, which is why the electrical current conducted via the respective circuit breaker is greater than ten times or twice its rated current. To prevent further damage and / or thermal malfunction from occurring with such a large amount of electrical energy, the electrical current must be interrupted relatively quickly by the circuit breaker. In the case of an overcurrent, on the other hand, the electrical current conducted by the circuit breaker is lower and can be, for example, up to three times or 1.4 times the rated current. Immediate interruption is therefore not necessary, at least if the overcurrent only lasts for a comparatively short period of time.Such an event occurs, for example, when one of the loads is switched on, especially if it contains electrical capacitances that are charged upon switching on. However, due to the increased electrical current, the load on the circuit breaker and the circuit is increased. To prevent damage, the electrical current supplied to the load via the circuit breaker is usually limited. Due to the current limitation, normal operation of the load is possible after the overcurrent has subsided.

[0006] To achieve current limitation, a semiconductor switch in the circuit breaker is typically used, through which the electrical current is conducted. This switch is then operated in current-limiting mode, also known as linear mode. Due to the limitation of the electrical current, an increased electrical voltage drops across the semiconductor switch, so that power loss in the semiconductor switch is increased. This leads to heating of the semiconductor switch, which can lead to destruction if linear operation is maintained for a comparatively long period of time. For this reason, it is provided, for example, that a temperature sensor is arranged in the semiconductor switch in the region of the junction layer, by means of which the junction temperature is detected.If this exceeds a critical value, the current flow is completely interrupted, preventing further heating and preventing destruction of the semiconductor switch. However, such semiconductor switches are comparatively expensive. Alternatively, the housing temperature of the semiconductor switch can be measured using a suitable temperature sensor. This reduces manufacturing costs. However, the temperature prevailing in the housing area can deviate relatively significantly from the junction temperature, which is crucial for the operational reliability of the semiconductor switch. Another alternative involves determining the junction temperature using a theoretical model and the operating data of the semiconductor switch. The Cauer or Forster model is typically used for this purpose. This model is used to describe an ideal semiconductor switch.Thus, a separate temperature sensor within the semiconductor switch is not required.

[0007] During operation, however, thermal instabilities may occur, particularly the so-called Spirito effect, in which an elevated temperature in certain areas of the semiconductor switch leads to increased electrical current flow, which increases power dissipation and thus increases heating. To prevent damage to the semiconductor switch, it is switched to a current-blocking state even when the junction temperature determined using the theoretical model has approached the critical value / temperature within a comparatively large tolerance value. In other words, a reduced operating range (SOA) of the semiconductor switch is selected for safety reasons. As a result, the semiconductor switch, and thus the associated load, is switched off more frequently.

[0008] The invention is based on the object of specifying a particularly suitable method for operating a circuit breaker and a particularly suitable circuit breaker, wherein advantageously a number of unnecessary trips is reduced and / or operational reliability is increased.

[0009] With regard to the method, this object is achieved according to the invention by the features of claim 1 and with regard to the circuit breaker by the features of claim 6. Advantageous further developments and refinements are the subject of the respective subclaims.

[0010] The method serves to operate a circuit breaker. The circuit breaker serves in particular to protect a device, and the circuit breaker is, for example, a device circuit breaker. Alternatively or in combination with this, the circuit breaker serves to protect a line and is thus a line circuit breaker. At least preferably, the circuit breaker is suitable, in particular provided and configured, to be introduced into an electrical circuit and to interrupt a current flow through it in the event of a fault. Preferably, the circuit breaker can be mounted in a power distributor or, in the mounted state, is a component of the power distributor. The circuit breaker is particularly suitable, expediently provided and configured for this purpose. For example, the circuit breaker is provided for switching an electrical direct current and / or a electrical direct voltage.Alternatively, the circuit breaker is intended to interrupt an alternating electrical current / voltage. For example, the current-carrying capacity / rated current of the circuit breaker is greater than 0.1 A, 0.5 A, 1 A, 2 A, or 5 A. In particular, the rated current of the circuit breaker is less than 50 A, 40 A, 30 A, 20 A, 15 A, or 10 A. In particular, the maximum electrical voltage that can be switched by the circuit breaker is greater than 2 V, 5 V, 10 V, or 20 V. In particular, the maximum electrical voltage that can be switched by the circuit breaker is less than 100 V, 80 V, 70 V, 60 V, or 50 V.

[0011] The circuit breaker has a switching path. In particular, the circuit breaker comprises two terminals, between which a current path is formed, which is also referred to as the main current path, and by means of which, for example, the switching path is formed. At least the current path expediently comprises the switching path. The two terminals serve, in particular, to connect the circuit breaker to other components of a circuit, which is protected, in particular, by the circuit breaker. The circuit breaker has at least two different states, namely an electrically conductive (closed) and an electrically non-conductive (open) state.

[0012] The switching path comprises a semiconductor switch. In particular, the semiconductor switch conducts the electrical current conducted through the circuit breaker, provided the circuit breaker is in the electrically conductive state. For example, the semiconductor switch is a power semiconductor switch. For example, the semiconductor switch is a field-effect transistor, such as a MOSFET, a bipolar transistor, a thyristor, a GTO, or an IGBT. The semiconductor switch suitably has a control input to which an electrical potential can be applied, thereby changing the conductivity of the semiconductor switch.

[0013] The circuit breaker further comprises a current sensor which is assigned to the switching path. The assignment is in particular such that the electrical current conducted via the switching path can be measured by means of the current sensor. The current sensor is preferably used to detect the electrical current conducted via the switching path and thus also the electrical current conducted by means of the semiconductor switch. For example, the current sensor is a component of the switching path and is thus incorporated into it. In this case, the current sensor is expediently connected electrically in series with the semiconductor switch. For example, the current sensor is a shunt or at least comprises one. Alternatively, the current sensor comprises, for example, a Hall sensor or a reed relay. In this case, the current sensor is in particular not a component of the switching path.The circuit breaker expediently also comprises a voltage sensor, which can be used to detect, in particular measure, the electrical voltage drop across the semiconductor switch. Alternatively, the electrical voltage drop across the semiconductor switch can be detected, in particular based on other operating data of the semiconductor switch.

[0014] The method provides that the semiconductor switch is controlled such that it is operated in linear mode. In other words, the semiconductor switch is not operated in saturation, and by changing the electrical potential applied to the control input, preferably the gate, the electrical current conducted through the semiconductor switch, in particular between source and drain, is changed, in particular continuously. In other words, the semiconductor switch is operated in current-limiting mode, and this current is not fully turned on. In particular, when the circuit breaker is electrically conductive, the semiconductor switch is operated in linear mode. Suitably, the semiconductor switch is only operated in linear mode when a specific condition is met. In particular, this condition is met when an overcurrent is present.This is the case, for example, if the electrical current conducted through the circuit breaker is greater than twice the rated current of the circuit breaker. However, the electrical current conducted through the circuit breaker is always less than ten or five times the rated current. For example, the current sensor is used to determine whether the specific condition is met. Preferably, in linear operation, the electrical current conducted by the semiconductor switch is controlled or regulated, for which purpose the current sensor is used in particular. In other words, the electrical potential applied to the control input of the semiconductor switch is changed depending on the electrical current measured by the current sensor.

[0015] If, however, the condition is not met, the semiconductor switch is preferably operated in full conduction mode, specifically in saturation mode. If another condition is met, for example, a short-circuit current is present, the semiconductor switch is expediently controlled in such a way that it is current-blocking, thus preventing any electrical current flow through the circuit breaker. A short-circuit current is present when the electrical current flowing through the circuit breaker is greater than ten times its rated current.

[0016] According to the method, a power loss occurring in the semiconductor switch during linear operation is determined, as well as the limiting layer temperature of the semiconductor switch resulting from the power loss. A model is used for this purpose, in particular. In other words, the junction temperature is determined, i.e., in particular, estimated, using the model based on the power loss. The so-called Cauer or Forster model is preferably used for this purpose. In summary, the resulting junction temperature is expediently determined using the model based on the power loss. In this case, in particular, the power loss leads to an assumed heating during a time window, which preferably decays exponentially or at least has an exponential curve.The period of linear operation is divided into several such time windows, and the respective heatings are in particular added up or integrated.

[0017] The determined junction temperature is compared with a limit value, which is either constant or dependent on variables. The limit value is preferably statically specified and, in particular, selected depending on the semiconductor switch. For example, the limit value is specified based on a data sheet for the semiconductor switch. For example, it is possible to change the limit value during operation or at least during installation / maintenance of the circuit breaker, for example by programming the circuit breaker and / or applying a specific reference voltage to a corresponding input of the circuit breaker.

[0018] Depending on the comparison, the control of the semiconductor switch is changed. In particular, if the determined junction temperature is greater than the limit value, a different control is used than if the junction temperature is less than the limit value. Thus, if the junction temperature is greater than the limit value during linear operation due to power loss, the control of the semiconductor switch is expediently changed, preferably in such a way that further heating is prevented. This prevents destruction of the semiconductor switch.

[0019] In this method, the power loss is determined based on the electrical current conducted by the semiconductor switch and the electrical voltage drop across the semiconductor switch. For this purpose, the electrical current conducted across the switching path and thus across the semiconductor switch is measured using a current sensor and used to determine the power loss. A voltage sensor is also expediently present, by means of which the electrical voltage drop across the semiconductor switch is measured. Furthermore, the power loss is determined based on a factor. The factor itself, in turn, depends on the duration of the linear operation. In other words, the factor in particular changes if the linear operation lasts for a longer period. As a result, the power loss changes, even if the electrical current conducted and the voltage drop have not changed.Alternatively, or in combination with this, the factor depends on the electrical voltage drop across the semiconductor switch. Preferably, the factor also depends on further parameters, such as a housing temperature of the semiconductor switch. For example, the power loss is calculated, in particular using a formula. Suitably, the power loss is a product of the conducted electrical current, the dropped electrical voltage, and the factor. The power loss expediently comprises at least the product. Alternatively, the power loss is stored, for example, in a further characteristic map as a function of at least these three variables.

[0020] The method therefore does not use, or at least not only, the actual power loss, but rather the power loss altered / falsified due to the factor. Consequently, the factor is also used to determine the junction temperature, which is subsequently altered or falsified due to the factor. Consequently, it is still possible to use existing models, such as the Cauer or Forster model, to continue to determine, in particular calculate, the junction temperature based on the power loss. The junction temperature can be determined using equivalent RC networks, and it is only necessary to supplement an existing circuit breaker design with the use of the factor, whereas the other components do not need to be changed.This factor takes into account thermal characteristics that distinguish the semiconductor switch from an ideal semiconductor switch and / or that arise from the design / construction of the semiconductor switch. This makes it possible to select the limit value relatively close to the safe operating limit (SOA), in particular the critical value, above which destruction of the semiconductor switch is to be expected due to its material properties, and / or which is specified, for example, in a data sheet for the semiconductor switch. This reduces the number of unnecessary trips while still maintaining or increasing operational reliability.

[0021] To determine the junction temperature, a formula is used which includes or consists of the following expression:

[0022] Here, P represents the current power dissipation, Ri the thermal resistance, and ti the thermal time constant. Ri and ti are specified by the semiconductor switch manufacturer.

[0023] For example, the factor is only changed for certain time periods, so that during linear operation the factor assumes the same value several times in succession. However, the factor is particularly preferably a monotonic function of the duration. For example, the function is monotonically decreasing, particularly if additional cooling is present. However, the function is particularly preferably monotonically increasing, so that the factor is selected such that it is increased for a longer duration. As a result, the longer the linear operation lasts, the greater the assumed power loss. Therefore, the resulting junction temperature (effective junction temperature) is higher than the actual junction temperature, which in particular leads to an earlier change in the control. As a result, the factor is selected such that thermal instabilities are taken into account, preferably the so-called Spirito effect.At the very least, the factor appropriately takes into account self-reinforcing effects that lead to excessive heating, for example, either completely or only locally, such as in so-called "hot spots." In other words, the factor preferably takes into account the fact that the heating of the semiconductor switch is not uniform, for example due to its design, but that there are regions where excessive heating occurs during linear operation. This prevents destruction of the semiconductor switch, while the limit value can still be selected to be comparatively high.

[0024] For example, the factor is calculated using a formula, in particular using one or more equations that are, for example, linear or non-linear. This reduces hardware requirements in particular. If the factor depends on the electrical voltage drop across the semiconductor switch, there is expediently a linear relationship between the two. In particular, the factor or at least the further value to be multiplied by the value is the sum of a constant, such as 0.6, and the product of another constant, such as 0.0144, and the electrical voltage drop across the semiconductor switch. In particular, a case distinction is made here, and the sum is only used if the electrical voltage drop across the semiconductor switch is greater than a certain limit, such as 27.8 V. Otherwise, 1 is used as the factor or further value.

[0025] For example, the factor is determined using an electrical network, e.g. RC elements. For example, the factor is continuous or only has discrete values / steps. Particularly preferably, the factor is read from a characteristic map. This enables a comparatively quick determination of the factor and thus the junction temperature. The characteristic map was determined, for example, based on a measurement, so that the use of a comparatively complicated equation, which may also contain approximations, is not necessary. Consequently, accuracy is improved and the limit value can be selected to be comparatively large. In the characteristic map, the factor is stored, for example, only as a function of the duration of linear operation or the electrical voltage across the semiconductor switch, so that the characteristic map is formed using a table.Alternatively, the factor is stored as a function of both the duration of linear operation and the electrical voltage across the semiconductor switch and / or as a function of other variables, so that the characteristic map is designed multidimensionally. For example, the factor is continuous or, particularly preferably, has only discrete, different values / levels. The number of different values / levels is, for example, 2 or more, for example 3 or 4. The number of different levels, i.e., the possible variations of the factor, is expediently less than 10. This reduces the effort required. For example, the factor is 1 for a duration of less than 10 ms, 2 for a duration between 10 ms and 100 ms, and 3 for a duration longer than that.

[0026] For example, the factor is first read from the characteristic map, and then the junction temperature and / or the power loss are read from another characteristic map, where the variables are the factor, the electrical current, and the electrical voltage. Alternatively, the characteristic map in which the factor and are stored is integrated into the additional characteristic map, so that the junction temperature / power loss is read from the additional characteristic map, in which it is stored as a function of the duration, the electrical voltage, and the electrical current. In other words, the factor is not determined explicitly. This reduces the effort involved.

[0027] For example, the junction temperature is determined solely based on the power loss. However, it is particularly preferred to also consider the housing temperature of the semiconductor switch when determining the junction temperature. The housing temperature is expediently measured using a temperature sensor, which is arranged, for example, outside or inside a housing of the semiconductor switch. For example, the junction temperature determined based on the power loss is corrected using the housing temperature, or the latter is already used as an offset. If the additional characteristic map is used, the junction temperature / power loss as a function of the housing temperature is expediently stored there.By using the housing temperature, a specific application area of ​​the semiconductor switch is taken into account, for example, whether it is used in a relatively cold or warm environment, or whether it has already warmed up due to previous linear operation or other operation. Manufacturing the semiconductor switch with this type of temperature sensor is comparatively inexpensive, thus improving the accuracy of determining the junction temperature with only a slightly increased manufacturing cost.

[0028] To change the control, the semiconductor switch is preferably put into a current-reduced state. For example, the electrical current carried by the semiconductor switch is reduced, in particular if the determined junction temperature is greater than the limit value. In this case, an electrical current continues to be carried by the semiconductor switch, for example. This thus enables continued operation, albeit at a reduced level, of the load protected by the circuit breaker, wherein any resulting heating of the semiconductor switch is reduced. However, to change the control, the semiconductor switch is preferably put into an electrically blocking state. In particular, the semiconductor switch is put into the electrically blocking state if the determined junction temperature is greater than the limit value.As a result, further heating of the semiconductor switch is excluded and the limit value can be chosen comparatively large.

[0029] The circuit breaker has a switching path that includes a semiconductor switch. The switching path expediently extends between two terminals of the circuit breaker that are suitable, in particular provided and configured, for connection to an electrical circuit. If the circuit breaker is in an electrically conductive state, an electrical current is conducted via the switching path. The electrical current is then conducted via the semiconductor switch. In particular, the semiconductor switch comprises a control input via which the electrical conductivity of the semiconductor switch can be changed. Depending on an electrical potential applied to the control input, the semiconductor switch is in an electrically blocking, an electrically fully conductive (saturation operation), or a current-limiting operating state.In the fully electrically conductive operating state (saturation mode), the semiconductor switch, and thus the switching path, exhibits a comparatively low electrical resistance. In the current-limiting operating state (linear mode), however, the electrical resistance is increased, although an electrical current still flows. The magnitude of the flowing electrical current depends on the electrical potential applied to the control input, with a linear / proportional relationship, for example.

[0030] The circuit breaker also has a current sensor associated with the switching path. The current sensor is arranged in such a way that it can measure the electrical current flowing through the switching path.

[0031] The circuit breaker is operated according to a method in which the semiconductor switch is controlled such that it operates in linear mode, expediently when at least one specific condition is met. This condition is met, for example, when an overcurrent is present. During linear operation, the power loss occurring in the semiconductor switch and the resulting junction temperature of the semiconductor switch are determined. The determined junction temperature is compared with a limit value, and the control of the semiconductor switch is changed depending on the comparison.The power loss is determined based on the electrical current conducted by the semiconductor switch and the voltage drop across the semiconductor switch, as well as a factor that depends on the duration of linear operation, i.e., how long the semiconductor switch has been actuated. Alternatively, or in combination with this, the factor depends on the voltage drop across the semiconductor switch. In particular, the factor also depends on other parameters, operating variables, and / or operating data of the circuit breaker.

[0032] In particular, the circuit breaker has a control unit that is suitable, in particular provided and configured, for carrying out the method at least partially or completely. The semiconductor switch is expediently controlled and / or the current sensor is read by means of the control unit. In particular, the control unit is configured analog or, for example, at least partially digital. Suitably, the circuit breaker comprises a microcontroller, thus facilitating the determination of the junction temperature.

[0033] The circuit breaker is preferably also suitable, in particular provided and configured, for the semiconductor switch to be electrically blocked in the event of a short-circuit current and / or another fault. The control unit is suitably adapted accordingly for this purpose. The fault / short-circuit current is expediently detected by the current sensor or another sensor, and the semiconductor switch is actuated accordingly, preferably by the control unit. If, however, it is determined by the current sensor or another sensor or due to other circumstances that there is no short-circuit current but rather an overcurrent, the semiconductor switch is in particular switched to linear operation and the junction temperature is determined accordingly.

[0034] The further developments and advantages explained in connection with the process can also be transferred to the circuit breaker and vice versa.

[0035] An embodiment of the invention is explained in more detail below with reference to a drawing. In the drawings:

[0036] Fig. 1 is a block diagram of a circuit with a circuit breaker, and,

[0037] Fig. 2 shows a method for operating the circuit breaker.

[0038] Corresponding parts are provided with the same reference numerals in all figures.

[0039] Figure 1 shows a block diagram of an electrical circuit 2. The electrical circuit 2 has a DC voltage source 4, which provides a DC voltage of 24 V. This voltage is applied between a supply output 6 and ground 8. The voltage source is a component of a power distributor (not shown in detail). The electrical circuit 2 also has a load 10, which is electrically connected to ground 8 on the one hand and to the supply output 6 on the other hand by means of a line 12 and a circuit breaker 14.The circuit breaker 14 is also a component of the power distribution and serves to protect the line 12 and the load 10 against destruction due to an increased electrical current, namely either an overcurrent, in which the current between the DC voltage source 4 and the load 10 is up to three times the rated current for which the circuit breaker 14 is designed, or a short-circuit current, in which, for example, a low-resistance connection exists between the line 12 and ground 8.

[0040] The circuit breaker 14 has a switching path 16 extending between two terminals 18 of the circuit breaker 14, which are incorporated into a housing 20 of the circuit breaker 14. The switching path 16 is arranged entirely within the housing 20, and one of the terminals 18 is directed toward the supply output 6 and the other terminal 18 is directed toward the line 12.

[0041] The switching path 16 has a semiconductor switch 22, which is configured as a MOSFET and is connected in series with a current sensor 24. This is thus a component of the switching path 16 and is therefore associated with it. The current sensor 24 is configured as a shunt, and by means of it, the electrical current conducted between the terminals 18, and therefore also the current via the current sensor 24, can be measured.

[0042] The current sensor 24 is signal-connected to a control unit 26 of the circuit breaker, which is designed, for example, as a common structural unit or comprises several, for example distributed, structural units. At least during operation, the measurement data determined by the current sensor 24 are read out by the control unit 26. For this purpose, the control unit 26 has a microcontroller 28 and electrical and / or electronic components not shown in detail. The semiconductor switch 22 is also actuated by the control unit 26, for which purpose a corresponding electrical potential is applied to a control input 30 of the semiconductor switch 22 by the control unit 26. The electrical resistance provided by the semiconductor switch 22 depends on the level of the electrical potential and is either high-resistance or low-resistance, or has an intermediate value.If the semiconductor switch 22 has a high resistance, it is electrically blocked, and no electrical current flow is possible between the terminals 18. In this state, the protective switch 14 is open. If the semiconductor switch 22 has a low resistance, the ohmic resistance provided by the semiconductor switch 22 is essentially negligible or at least comparatively low. This allows a substantially uninterrupted electrical current flow between the two terminals 18, and the protective switch 14 is closed. To achieve this state, the semiconductor switch 22 is operated in saturation and is thus fully controlled.

[0043] If an intermediate value of the ohmic resistance is provided by the semiconductor switch 22, it is possible to limit the electrical current conducted between the terminals 18 by means of the semiconductor switch 22. In this case, an increased electrical voltage is generated across the semiconductor switch 22. There is a linear relationship between the electrical current conducted via the semiconductor switch 22 and the electrical potential present at the control input 30. Such operation of the semiconductor switch 22 is referred to as linear operation. Due to the increased applied electrical voltage and the conducted electrical current, electrical losses occur at the semiconductor switch 22, which lead to heating.

[0044] The control unit 26 detects the electrical voltage drop across the semiconductor switch 22. For this purpose, the control unit 26 comprises a voltage sensor (not shown in detail) connected in parallel with the semiconductor switch 22. The semiconductor switch 22 also has a temperature sensor 32, which is also read by the control unit 26. This temperature sensor is located in the immediate vicinity of a housing (not shown in detail) of the semiconductor switch 22 and is thus thermally connected to it.

[0045] Figure 2 shows a method 34 for operating the circuit breaker 14, which is carried out at least partially by the control unit 26. The circuit breaker 14 is also operated in parallel or alternately according to methods not shown in detail, so that the method 34 is only used for part of the operation of the circuit breaker 14. The method 34 is started in a first work step 36. The first work step 34 is carried out when a specific condition is met, namely that an overcurrent prevails, i.e. that an electrical current is conducted across the switching path 16 that is greater than the rated current but less than three times the rated current. The flowing electrical current is measured by the current sensor 24 and read out by the control unit 26, and it is checked whether the overcurrent prevails, i.e. whether the specific condition is met.

[0046] The overcurrent occurs, for example, when the current is applied to the load 10 and this load has electrical capacitances that are initially charged. In this case, the overcurrent is rather small and prevails for a short period of time, namely until the electrical capacitances are at least partially charged. If, on the other hand, the electrical current is greater than the overcurrent, a different condition is met and a different method is carried out in which the semiconductor switch 22 is essentially immediately put into the electrically blocking state. In method 34, on the other hand, the semiconductor switch 22 is controlled in the first work step such that it is operated in linear mode. In this mode, the electrical current flow via the semiconductor switch 22 is still possible, but the level of the electrical current is limited.For example, it is constant or depends on the duration of the linear operation.

[0047] For control, a corresponding electrical potential is applied to the control input 30 by means of the control unit 26, with the current sensor 24 detecting the level of the electrical current conducted through the switching path 16. By tracking the level of the applied electrical potential, control to the desired level of the electrical current is possible.

[0048] In a subsequent second work step 38, which is thus carried out during the linear operation of the semiconductor switch 22, the electrical current conducted through the semiconductor switch 22 is determined. This corresponds to the electrical current conducted through the current sensor 24 and is measured by means of the latter. In addition, the electrical voltage dropping across the semiconductor switch 22 is recorded, namely directly by means of the appropriately equipped control unit 26. In addition, a housing temperature of the semiconductor switch 22 is measured by means of the temperature sensor 32 and read out by the control unit 26. Thus, the control unit 26 has information about the currently conducted electrical current, the applied electrical voltage, and the housing temperature. Since the semiconductor switch 22 is also controlled by the control unit 26, it also has information about the duration of the linear operation, i.e. how long the linear operation has already lasted.

[0049] In addition, a factor is determined by the control unit 26. For this purpose, the factor is read out from a characteristic map 40 present in the control unit 26 by the microcontroller 28. In summary, the factor is read out from the stored characteristic map 40. In the example shown, the characteristic map 40 is a table in which the factor is stored as a function of the duration of the linear operation. However, it is also possible for the characteristic map 40 to have an additional dimension, and the factor is also stored as a function of the electrical voltage currently dropping across the semiconductor switch 22. In the example, three different, discrete values ​​for the factor are stored in the characteristic map 40. For a duration of less than 10 ms, the factor is 1. Above this and up to a duration of 100 ms, the factor is 2, and for a longer duration, the factor is 3.In a further development, the factor is read from the characteristic map 40 and then corrected depending on the electrical voltage drop across the semiconductor switch. The factor remains the same if the electrical voltage drop across the semiconductor switch is less than 27.8V. If the electrical voltage is greater than 27.8V, the value read from the characteristic map 40 is multiplied by another value, and the product corresponds to the factor. The further value corresponds to the sum of 0.6 and the product of 0.0144 and the electrical voltage drop across the semiconductor switch. The electrical current currently conducted via the semiconductor switch 22, the electrical voltage currently drop across the semiconductor switch 22 and the currently applicable factor are multiplied by the microcontroller 28 to form a current (effective) power loss 42.

[0050] In a subsequent third work step 44, a junction temperature 46 is determined based on the power loss 42 and the housing temperature of the semiconductor switch 22, which was also determined in the second work step 38. For this purpose, a model 48 stored in the control unit 26, namely either the Cauer or Forster model, is used, and the necessary calculations are carried out using the microcontroller 28. The housing temperature of the semiconductor switch 22 serves as an offset, so that it is also taken into account when determining the junction temperature 46. In summary, the input values ​​for the model 48 are the (current) power loss 42, the housing temperature, and a current time. The model 48 preferably has parameters that are, however, static and that are adapted to the semiconductor switch 22 used.

[0051] To determine the junction temperature 46, the following formula is used:

[0052] Here, T describes Sperr ^(t) is the junction temperature 46, UDS is the electrical voltage drop across the semiconductor switch 22, and io is the electrical current conducted by the semiconductor switch (22). f(t, u DS is the factor that depends on UDS and the duration of linear operation (duration of current limitation). f(t,u DS ■ u DS t) ■ i D t) ■ is the power loss 42, Rth the thermal

[0053] resistance and Cth the thermal capacitance. T Gehäuse is the determined housing temperature of the semiconductor switch 22.

[0054] In a subsequent fourth work step 50, the determined junction temperature 46, which results from the power loss 42, is compared with a limit value 52 which depends on the material of the semiconductor switch 22 and which is fixed. The limit value 52 is slightly below the temperature at which destruction of the semiconductor switch 22 is to be expected, for example 2 °C below. If the determined junction temperature 46 is greater than the limit value 52, a fifth work step 54 is carried out in which the control of the semiconductor switch 22 is changed. For this purpose, it is placed in an electrically blocking state so that it has a high resistance. For this purpose, the electrical potential applied to the control input 30 is selected accordingly using the control unit 26. As a result, the electrical current flow via the switching path 16 is completely interrupted, thus shutting down the load 10.As a result, semiconductor switch 22 no longer experiences losses, and it also no longer heats up. This prevents destruction of semiconductor switch 22 and circuit breaker 14.

[0055] If, however, the junction temperature 46 is lower than the limit value 52, the second work step 38 is carried out again and the then prevailing electrical current, the falling electrical voltage and the factor, and based on these, the current power loss 42, are determined. The third and fourth work steps 44, 50 are also carried out again. If the linear operation lasts longer, i.e. if the second work step 38 has already been carried out several times, a different value is used for the factor. As soon as the duration is greater than 10 ms, 2 is used as the factor or the value to be multiplied by the further value, and if 100 ms have already elapsed since the first work step 36 was carried out, 3 is used as the factor or value. Consequently, the factor is selected such that it is increased with a longer duration of linear operation.Due to the then increased factor, the power loss 42 determined thereby is falsified and tends to be increased. As a result, the junction temperature 46, which is determined in the subsequent third work step 44, is also greater than the actual junction temperature or at least the junction temperature that would be determined using the model 48 if the actual / unadjusted power loss were used. As a result, the fifth work step 54 tends to be carried out earlier. However, it is possible to select a comparatively large limit value 20, whereby thermal peculiarities such as local excessive heating are avoided due to the factor. One such thermal peculiarity / instability is, for example, the Spirito effect, in which self-reinforcing effects occur in a specific operating range of the semiconductor switch 22.Thus, heating leads to increased electrical current flow, which leads to further / greater heating. As a result, and due to the design of the semiconductor switch 22, which is configured as a MOSFET, hot spots can arise whose temperature is significantly higher than the temperature of the rest of the junction of the semiconductor switch 22. In other words, the temperature of the junction is not uniform, which is taken into account by the factor.

[0056] Thus, despite the comparatively high threshold, no destruction occurs in method 34, and the high threshold reduces the frequency with which the fifth step 50 is performed. It is possible to use the existing models 48, and the thermal characteristics are taken into account only by means of the factor, so that an existing design of the circuit breaker 14 only needs to be modified slightly.

[0057] If the electrical current drops again to the nominal value or a lower value during the execution of method 34, method 34 is terminated, wherein semiconductor switch 22 is again placed in the fully conductive state. In other words, semiconductor switch 22 is again operated in saturation mode, and method 34 is terminated. The invention is not limited to the exemplary embodiment described above. Rather, other variants of the invention can also be derived therefrom by a person skilled in the art without departing from the subject matter of the invention. In particular, all individual features described in connection with the exemplary embodiment can also be combined with one another in other ways without departing from the subject matter of the invention.

[0058] List of reference symbols

[0059] 2 circuits

[0060] 4 DC voltage source

[0061] 6 Supply output

[0062] 8 Mass

[0063] 10 Last

[0064] 12 Line

[0065] 14 circuit breakers

[0066] 16 switching distance

[0067] 18 connection

[0068] 20 housings

[0069] 22 semiconductor switches

[0070] 24 Current sensor

[0071] 26 Control unit

[0072] 28 microcontrollers

[0073] 30 control input

[0074] 32 Temperature sensor

[0075] 34 procedures

[0076] 36 first work step

[0077] 38 second step

[0078] 40 map

[0079] 42 Power loss

[0080] 44 third step

[0081] 46 Junction temperature

[0082] 48 Model

[0083] 50 fourth step

[0084] 52 Limit

[0085] 54 fifth step

Claims

Claims 1. Method (34) for operating a circuit breaker (14) with a switching path (16) which has a semiconductor switch (22) and to which a current sensor (24) is assigned, in which - the semiconductor switch (22) is controlled in such a way that it is operated in a linear mode, - a power loss (42) occurring in the semiconductor switch (22) during linear operation and a resulting junction temperature (46) are determined, - the determined junction temperature (46) is compared with a limit value (52), and - depending on the comparison, a control of the semiconductor switch (22) is changed, wherein the power loss (42) is determined on the basis of the electrical current carried by means of the semiconductor switch (22) and the electrical voltage dropping across the semiconductor switch (22) as well as a factor which is dependent on the duration of the linear operation and / or the electrical voltage dropping across the semiconductor switch (22).

2. Method (34) according to claim 1, characterized in that the factor is selected such that it is increased for a longer duration.

3. Method (34) according to claim 1 or 2, characterized in that the factor is read out from a stored characteristic map (40).

4. Method (34) according to one of claims 1 to 3, characterized in that a housing temperature of the semiconductor switch (22) is taken into account when determining the junction temperature (46).

5. Method (34) according to one of claims 1 to 4, characterized in that in order to change the control, the semiconductor switch (22) is placed in an electrically blocking state.

6. A circuit breaker (14) comprising a switching path (16) having a semiconductor switch (22) and associated with a current sensor (24), and operated according to a method (34) according to any one of claims 1 to 5.

Images