Device for protecting a power semiconductor

Abstract

Device (1) for protecting a power semiconductor, comprising • a current meter (U R ), a voltmeter (U SW ), and a switch (SW), • where the current meter (U R ) measures a current (I) through the device (1), wherein the switch (SW) is suitable to interrupt a current flow through the device (1), wherein the voltmeter (U) SW ) measures a voltage across the switch (SW) in the open state, • where, if the ammeter (U R ) measures a current (I) that meets a first current threshold (I) ref ) exceeds, the switch (SW) will remain on for a predetermined time (t) aus ) is controlled in such a way that the current flow is interrupted, within a predetermined time (t aus ) using the voltmeter (U SW ) a voltage is measured across the open switch (SW), • where, in the case of repeated switching, the voltage (U) SW) above the open switch (SW) it is detected that a short circuit is present, and the switch (SW) is permanently controlled in such a way that the current flow is interrupted. • wherein, if after opening the switch (SW) a voltage (U) is present at the switch (SW) SW ) is applied, which is lower than the mains voltage, so that it is recognized that there is no short circuit.

Description

[0001] The invention relates to a device for protecting a power semiconductor.

[0002] Numerous devices for protecting a power semiconductor are known from the prior art.

[0003] Distinguishing between real and virtual short circuits proves to be a major problem. Virtual short circuits arise from, for example, capacitive loads, which behave like a short circuit, especially when switched on. This means that considerable currents flow until the capacitive load is sufficiently charged.

[0004] Therefore, attempts have been made repeatedly in the past to detect these virtual short circuits.

[0005] For example, an electronic safety device is known from DE 10 2012 103 551 A1 which switches off a controllable switch depending on an overcurrent and switches it back on depending on the voltage change at the output.

[0006] A disadvantage of these devices is that they initially shut down in response to an overcurrent, and only allow reconnection after a certain period, depending on a reconnection condition. This often results in an unnecessarily long time required to charge a capacitive load. Furthermore, detection is comparatively complex, as it requires the storage of "historical" values. Another common problem with such arrangements is that, after the switch opens, the capacitive load is discharged by a parallel resistive load, potentially leading the detection system to incorrectly identify a capacitive (partial) load requiring charging as a short circuit. In particular, a device claimed in this context proves problematic with regard to switching on a capacitive load, as the switch-off condition is already met upon activation, thus eliminating the need for a historical voltage reference.

[0007] Furthermore, the previously known solution does not allow for a reliable statement about the actual state of the switch, as only the load side is considered.

[0008] German patent application DE 101 17 892 A1, which describes a limiting circuit against overload of a power supply, is known from the prior art. In this circuit, a switching signal is generated by a voltmeter across the switching element. This activates a variable resistor.

[0009] Furthermore, US patent application US 2012 / 0113555A1 is known from the prior art. This application describes an interruption device. The document aims to distinguish between a permanent fault and a temporary fault based on variations between successive cycles.

[0010] Furthermore, it is known from the German patent application DE 10 2007 006 564 A1 that the measured phase can also be used as a shutdown criterion.

[0011] Furthermore, a resettable circuit protection device is known from German patent application DE 10 2006 052 135 A1, in which an energy conversion can also be determined in the event of an excessive increase in current and an (indirectly measured) excessive increase in temperature.

[0012] The invention is based on the objective of creating improved and cost-effective devices for detecting a short circuit, which avoids one or more disadvantages of the prior art.

[0013] The problem is solved according to the invention by the features of the independent claim. Advantageous embodiments of the invention are specified in the dependent claims.

[0014] The invention is explained in more detail below with reference to the attached drawing and preferred embodiments.

[0015] They show Fig. 1 an overview of embodiments according to the invention, and Fig. 2 a flowchart of embodiments of the invention.

[0016] The invention will now be described in more detail with reference to the figure. It should be noted that different aspects will be described, each of which can be used individually or in combination. That is, each aspect can be used with different embodiments of the invention unless explicitly presented as a pure alternative.

[0017] Furthermore, for the sake of simplicity, reference will generally be made to only one entity at a time. Unless explicitly stated otherwise, the invention may also include several of the entities concerned. Therefore, the use of the words "a," "an," and "one" should only be understood as an indication that at least one entity is used in a simple embodiment.

[0018] In one embodiment of the invention, which is exemplified in the Fig. 1, a device 1 for protecting a power semiconductor includes a current meter U R , a voltmeter U SW , and a switch SW. For better clarity, device 1 in the figure is surrounded by a dashed-dotted frame.

[0019] The power semiconductor is not necessarily shown in the figures. That is, the power semiconductor can be another (not shown) element, or the switch SW itself can be provided by a power semiconductor, for example a MOSFET transistor (metal-oxide-semiconductor field-effect transistor). However, the invention is not limited to (switching) transistors.

[0020] The ammeter measures U R a current I through the device 1. In principle, any form of current measurement is possible for this purpose, be it by an induced magnetic field (e.g. Rogowski coil, Hall sensor) or by voltage measurement across a shunt resistor R or the like.

[0021] Switch SW is suitable for interrupting the current flow through device 1. In principle, any type of switch can be used here, e.g., a mechanical switch or an electrical switch, e.g., a controlled (field-effect) transistor.

[0022] Furthermore, the device uses a voltage meter U SW , which measures a voltage across switch SW. Since switch SW represents a short circuit in the idealized case, it is sufficient to measure voltages when switch SW is open, as the voltage should be zero when switch SW is closed.

[0023] Depending on the type of switch SW, it may also be designed so that, for example, the switch SW itself presents a low resistance, allowing the current measurement to use the closed switch as a shunt resistor R. Such a case can occur, for example, when using semiconductor switches (e.g., controlled (field-effect) transistors).

[0024] Therefore, the division described above should be understood functionally and does not limit the actual implementation.

[0025] If the ammeter U R measures a current I that meets a first current threshold I ref exceeds, i.e., I>I ref , the switch SW will be switched on for a predetermined time t ausThe device is controlled in such a way that the current flow through it is interrupted. As a rule, the current measurement is not performed immediately at the start of the current flow, but only after a predetermined minimum on-time t has elapsed. ein The current I is measured through the device.

[0026] Now, within this predetermined time t aus using the voltage meter U SW A voltage can be measured across the open switch SW.

[0027] This voltage can be stored. "Storage" here is to be understood generally and can mean both storing the current voltage in a capacitor and storing a measured value.

[0028] After the predetermined time t has elapsed aus The switch SW is switched on again, so that current flows again through the device 1 to the load V.

[0029] If switch SW is repeatedly activated as a result of an overcurrent, the voltage U can be used to determine the cause. SW It must be determined whether a genuine short circuit exists, which poses a serious danger.

[0030] This is simply possible because, in the case of a (repeated) shutdown and the presence of a capacitive load V, a voltage U is now present at the switch SW. SW The voltage applied is lower than the mains voltage. This is because a capacitive load is charged by the flowing current, so that after the switch SW is opened, the voltage charged by the current remains across the capacitive load V. In contrast, in a true short circuit, no charge is stored after the switch SW is opened, and therefore the voltage on the load side is 0 volts.

[0031] This can be determined, for example, with a simple differential voltage measurement, such as a Wheatstone bridge or a differentially driven operational amplifier or any other suitable device at switch SW.

[0032] If a true short circuit is detected, the switch (SW) will now be permanently controlled to interrupt the current flow.

[0033] Control can be either active ("opening") or passive ("not closing") and depends on the type of switch SW. That is, the switch SW can be configured as either a normally open or a normally closed switch.

[0034] In a further advantageous embodiment of the invention, it is also recognized that a temporary overcurrent occurs as a result of the charging of a capacitive load. When the voltmeter U SW during a temporary interruption t ausThe current flow measures a voltage at switch SW, which represents a first threshold value U. ref If the voltage exceeds the capacitive load V, it can be deduced that a capacitive load V is present. This is simply because, in the case of a (repeated) shutdown and the presence of a capacitive load V, a voltage U will now be present at switch SW. SW The voltage present is lower than the mains voltage. This is because a capacitive load is charged by the flowing current, so that after the switch SW is opened, the voltage charged by the current is still present across the capacitive load V. This voltage U V The load causes the voltage at the switch U to rise. SW now becomes lower than the mains voltage. Therefore, the presence of a difference (corresponding to U) can now be interpreted as follows: ref = 0 V) ​​or from the presence of a certain difference (corresponding to U) diff = U Netz - U V > U refA capacitive load V can be detected. Therefore, it is also possible to detect a load from the voltage across the switch U falling below a certain threshold. SW to detect the presence of a capacitive load.

[0035] In yet another advantageous embodiment of the invention, it is also possible that, for example, the predetermined time t aus The switching-off depends on the previously measured current IDh; for example, the current I is considerably larger than the current threshold I. ref , so the current can be switched off for a longer period of time, whereas, for example, with a current I that is only slightly greater than the current threshold I ref If this is the case, the current can be switched off for a shorter period of time. This prevents, for example, semiconductor switches from being unnecessarily overloaded, while at the same time ensuring a rapid charge of a capacitive load.

[0036] The present invention can also be designed as follows.

[0037] The device 1 for detecting an overcurrent includes an ammeter U R with a shunt resistor R, a calculation unit SE, and a switch SW.

[0038] The current meter U R The device measures a current through device 1 as a voltage across the shunt resistor R, and the switch SW is suitable for interrupting the current flow through device 1. As previously described, the shunt resistor can also be provided by the switch SW.

[0039] If the ammeter U R measures a current I that meets a first current threshold I ref If the value exceeds the set value, the switch SW will remain open for a predetermined time t. aus controlled in such a way that the current flow is interrupted.

[0040] The energy supplied to device 1 is determined from current values ​​measured at various times using the known shunt resistor R. The energy can be determined using current and voltage, as well as resistance and current, or resistance and voltage. If a maximum energy W is exceeded, the device can then be switched off. max The SW switch is permanently controlled in such a way that the current flow is interrupted.

[0041] A corresponding procedure can then be implemented, for example, as in Fig. 2 shown to be implemented.

[0042] In a first step, S50, the switch SW is activated so that current flows to the load V. After a predetermined time t has elapsed... ein In step S100, the current I is measured through device 1.

[0043] Additionally, it can optionally be provided that a voltage U is also applied after step S50. VThe voltage V is measured on the load side using a suitable voltmeter. This step can be performed during or after step S100.

[0044] In step S300, it is now checked whether a shutdown condition is met, namely whether the measured current I exceeds a first current threshold I. ref exceeds.

[0045] If this is not the case, the system returns to step S100 and executes the loop again.

[0046] If, however, the current I is greater than the first current threshold I ref , so in step S400 the switch SW is initially controlled temporarily so that it opens.

[0047] Next, the energy W that has flowed through device 1 or switch SW is determined. Depending on whether this determination also requires the determination of a voltage or not, the voltage U is then calculated. SWThe energy W is determined at the switch SW or from the known resistance R and the last measured current.

[0048] If the energy output W is greater than a certain maximum energy W max , then the SW switch will no longer be turned on and the procedure will end.

[0049] However, if the energy conversion W is less than (or equal to) a certain maximum energy W max so a time t aus Waited at step S675 and repeated the procedure from step S50.

[0050] It goes without saying that the conditions for exceeding the limit can be equivalently replaced by other conditions.

[0051] The steps and means mentioned above can be part of a control unit (SE) or executed by it. A control unit can be, for example, a microcontroller, an ASIC, or an FPGA.

[0052] The invention can be particularly advantageous, for example, in power supplies. These often contain high-performance switched-mode power supplies with controlled power semiconductors as switching elements.

[0053] A particular advantage of the invention is that it makes it possible to measure the actual load, in terms of energy input into the power semiconductor, based on the voltage drop across the power semiconductor as an embodiment of a switch SW. This measurement of energy input goes far beyond binary state detection (capacitive load, short circuit), because such detection must provide substantial safety margins.

[0054] Therefore, the presented invention also makes it possible, if necessary, to provide a current for a certain period of time, e.g. 10 ... 50 ms, that is considerably higher than the specified nominal current, e.g. 2 times to 10 times. Reference symbol list 1 Device U R electricity meter U SW Voltage meter SW switch I ref Current threshold t aus Predetermined time U ref threshold I Electricity R shunt resistance SE Calculation Unit V Last

Claims

[1] Device (1) for protecting a power semiconductor, comprising • a current meter (U R ), a voltmeter (U SW ), and a switch (SW), • where the current meter (U R ) measures a current (I) through the device (1), wherein the switch (SW) is suitable to interrupt a current flow through the device (1), wherein the voltmeter (U) SW ) measures a voltage across the switch (SW) in the open state, • where, if the current meter (U R ) measures a current (I) that meets a first current threshold (I) ref ) exceeds, the switch (SW) will remain on for a predetermined time (t) aus ) is controlled in such a way that the current flow is interrupted, within a predetermined time (t aus ) using the voltmeter (U SW ) a voltage is measured across the open switch (SW), • where, in the case of repeated switching, the voltage (U) SW) above the open switch (SW) it is detected that a short circuit is present, and the switch (SW) is permanently controlled in such a way that the current flow is interrupted. • wherein, if after opening the switch (SW) a voltage (U) is present at the switch (SW) SW ) is applied, which is lower than the mains voltage, so that it is recognized that there is no short circuit. [2] Device according to claim 1, wherein, when the voltage meter (U SW ) during a temporary interruption (t aus ) of the current flow measures a voltage that represents a first threshold value (U ref ) falls below a certain threshold, indicating that a capacitive load (V) is present. [3] Device according to claim 1 or 2, wherein the predetermined time (t aus ) depends on the previously measured current (I). [4] Device (1) for detecting an overcurrent, comprising • a current meter (U R) with a shunt resistor (R), a calculation unit (SE), and a switch (SW), • where the current meter (U R ) measures a current through the device (1) as a voltage across the shunt resistor (R), wherein the switch (SW) is suitable to interrupt a current flow through the device (1), • where, if the current meter (U R ) measures a current (I) that meets a first current threshold (I) ref ) exceeds, the switch (SW) will remain on for a predetermined time (t) aus ) is controlled in such a way that the current flow is interrupted, • wherein, from current values ​​measured at different times, the energy supplied to the device (1) is determined using the known shunt resistance (R) with the switch off, and if a maximum energy (W) is exceeded max ) the switch (SW) is permanently controlled in such a way that the current flow is interrupted.

Images