阅读说明：本技术 Afci自检 (AFCI self-check ) 是由 B·霍费尔 S·布鲁尔 C·法斯蒂伯 W·斯必泽 A·霍费尔 于 2019-09-26 设计创作，主要内容包括：为了识别在系统中的故障(F),该系统包括：变压器(4),变压器具有能与直流电压线路(DC+、DC-)串联的初级绕组(L1)和与初级绕组(L1)磁性耦联的次级绕组(L2)；以及评估单元(5),评估单元处理在次级绕组(L2)上的次级信号(i2),根据本发明,从次级信号(i2)中求得在至少一个频率(fn)下的信号分量(n(fn)),并且将信号分量(n(fn))的信号电平(P(fn))与信号阈值(P-min(fn))比较。当信号电平(P(fn))低于信号阈值(P-min(fn))时,探测到故障(F)。(In order to identify a fault (F) in a system, the system comprises: a transformer (4) having a primary winding (L1) which can be connected in series with the direct-current voltage line (DC +, DC-) and a secondary winding (L2) which is magnetically coupled to the primary winding (L1); and an evaluation unit (5) which processes the secondary signal (i2) at the secondary winding (L2), wherein according to the invention, a signal component (n) (fn)) at least one frequency (fn) is determined from the secondary signal (i2), and the signal level (P (fn)) of the signal component (n (fn)) is compared with a signal threshold (P _ min (fn)). When the signal level (P (fn)) is lower than the signal threshold (P _ min (fn)), a fault (F) is detected.)

1. A method for detecting a fault (F) in a system, the system comprising: a transformer (4) having a primary winding (L1) which can be connected in series with the direct-current voltage line (DC +, DC-) and a secondary winding (L2) which is magnetically coupled to the primary winding (L1); and an evaluation unit (5) which processes the secondary signal (i2) at the secondary winding (L2), characterized in that signal components (n (fn)) at least one frequency (fn) are determined from the secondary signal (i2) and the signal level (P (fn)) of the signal components (n (fn)) is compared with a signal threshold value (P _ min (fn)), and a fault (F) is detected in the system (1) when the signal level (P (fn)) falls below the signal threshold value (P _ min (fn)).

2. A method as claimed in claim 1, characterized in that at least one noise frequency of the fault-free system (1) which typically occurs on the secondary winding (L2) is taken into account as the at least one frequency (fn).

3. Method according to claim 2, characterized in that the noise level of the fault-free system (1) occurring at a typically occurring noise frequency is taken into account as a signal threshold (P _ min (fn)).

4. A method according to claim 3, characterized in that the signal threshold (P _ min (fn)) is determined in a range greater than zero and smaller than the noise level of the fault-free system (1) that occurs at the typically occurring noise frequency.

5. Method according to any of claims 1 to 4, characterized in that the signal components (n (fn)) are evaluated at a plurality of frequencies (fn) of a frequency band and the signal levels (P (fn)) at the plurality of frequencies (fn) are compared with a signal threshold (P _ min (fn)) respectively.

6. Method according to any of claims 1 to 4, characterized in that signal components (n (fn)) at a plurality of frequencies (fn) of a frequency band are determined, the signal levels (P (fn)) at a plurality of frequencies (fn) are averaged or median and compared with a signal threshold (P _ min (fn)).

7. A method according to claim 5 or 6, characterized in that the frequency band extends from 0Hz up to a limit frequency, preferably a limit frequency of 40 kHz.

8. The method according to any one of claims 1 to 7, characterized in that the signal component (n (fn)) is determined at the start-up of the system (1).

9. Method according to any of claims 1 to 8, characterized in that the evaluation unit (5) receives and evaluates the PLC receive signal (Rx (fx)) and/or the arc signal (arc (f _ arc)) transformed from the primary winding (L1) to the secondary winding (L2) by the transformer (4).

10. The method according to one of claims 1 to 9, characterized in that the PLC transmit signal (tx (fx)) is transmitted by the transmitting unit (3) to the secondary winding (L2) and is switched from the secondary winding (L2) onto the primary winding (L1).

11. Method according to one of claims 1 to 10, characterized in that a further signal component (n1(fn1)) at least one further frequency (fn1) is determined from the secondary signal (i2) and the signal level (P (fn1)) of the further signal component (n1(fn1)) is compared with an upper signal threshold (P _ max (fn1)) and a fault (F) in the system (1) is detected when the signal level (P (fn1)) is higher than the upper signal threshold (P _ max (fn 1)).

12. A method according to claim 11, characterized in that said further frequency (fn1) corresponds to said frequency (fn).

13. A system (1), the system comprising: a transformer (4) having a primary winding (L1) connectable in series with a direct voltage line (DC +, DC-), a secondary winding (L2) magnetically coupled with the primary winding (L1); and an evaluation unit (5) which is designed to process the secondary signal (i2) at the secondary winding (L2), characterized in that a fault evaluation unit (7) is provided which is connected to the secondary winding (L2) and is designed to determine from the secondary signal (i2) a signal component (n) (fn)) at least one frequency (fn), to compare the signal level (P (fn)) of the signal component (n (fn)) with a signal threshold value (P _ min (fn)), and to detect a fault (F) in the system (1) if the signal level (P (fn)) falls below the signal threshold value (P _ min (fn)).

14. The system (1) according to claim 13, wherein the evaluation unit (5) comprises an arc detection unit and/or a PLC receiving unit, which are designed to receive an arc signal (arc (f _ arc)) and/or a PLC receiving signal (rx (fx)) converted from the primary winding (L1) to the secondary winding (L2) by means of the transformer (4).

15. The system (1) according to claim 13, wherein the evaluation unit (5) is part of an arc detection unit and/or a PLC receiving unit which is designed to receive an arc signal (arc (f _ arc)) and/or a PLC receiving signal (rx (fx)) which is converted from the primary winding (L1) to the secondary winding (L2) by means of a transformer (4).

16. The system (1) according to one of claims 13 to 15, wherein a transmission unit (3) is provided, which is connected to the secondary winding (L2) of the transformer (4) and is designed to transmit the PLC transmission signal (tx (fx)) to the secondary winding (L2), and wherein the transformer (4) is designed to convert the PLC transmission signal (tx (fx)) from the secondary winding (L2) to the primary winding (L1).

17. The system (1) according to any one of claims 13 to 16, wherein the direct voltage line (DC +, DC-) is arranged for transmitting the direct voltage (Udc) of the at least one direct voltage source (2) to the at least one direct voltage bus (6).

18. The system (1) according to any one of claims 13 to 17, wherein said at least one direct voltage source (2) comprises at least one photovoltaic cell and said at least one direct voltage bus (6) comprises at least one inverter.

Technical Field

The invention relates to a method for detecting a fault in a system comprising: a transformer having a primary winding in series with the dc voltage line and a secondary winding magnetically coupled to the primary winding; and an evaluation unit which processes the secondary signal on the secondary winding. Furthermore, the invention relates to a system comprising: a transformer having a primary winding in series with a dc voltage line; a secondary winding magnetically coupled to the primary winding; and an evaluation unit which is designed to process the secondary signal on the secondary winding.

Background

In a dc voltage installation, a dc voltage is supplied by at least one dc voltage source and is transmitted to a dc voltage bus via a dc voltage line. As a direct voltage source, for example, a number of solar panels or solar cells of a photovoltaic installation, or a battery, can be provided. In this case, a direct voltage is generated on the solar panel or the solar cell as a function of the corresponding solar radiation. For example, the inverter may be used as a dc voltage bus. The inverter converts the direct voltage into an alternating voltage and can feed the alternating voltage into the grid or supply the generated alternating voltage to an electrical load, for example a motor or a battery.

In many systems, it is desirable to communicate with an existing dc voltage source and/or dc voltage bus. The communication signal sent by the control unit can be used, for example, to synchronize a plurality of individual solar panels. The communication signal can also be used to switch off the dc voltage source or other elements of the system, especially in the event of a fault situation.

The communication signal can thus be transmitted, for example, via a specially provided communication line or directly via an existing direct voltage line (by means of so-called power line communication, PLC). In the power line communication, a PLC signal known at a PLC frequency is modulated into a direct-current voltage for energy transmission as a base signal, and is transmitted. Thus, for coupling and/or decoupling of the PLC signal, a transformer can be used on the dc voltage line, wherein the primary winding is connected in series with the dc voltage line. The transformer converts an alternating primary signal flowing through the primary winding into an alternating secondary signal flowing in the secondary winding and vice versa. Furthermore, a PLC receiving unit or a PLC transmitting unit is connected with the secondary winding for receiving or transmitting a PLC signal. The further receiving or transmitting units can be located on a dc voltage source and/or a dc voltage bus connected to the dc voltage line.

The transformer can also be used to identify an arc when the arc detection unit is connected to the secondary winding of the transformer. If an arc occurs, the arc signal is converted by the transformer from the primary winding to a secondary signal on the secondary winding as part of the primary signal. The arc detection unit analyzes the secondary signal and identifies signal components with a certain arc frequency or signal components with different arc frequencies, which are converted to the arc signal on the secondary side by the transformer. From this, the occurrence of an arc can be inferred. When the arc signal on the secondary side differs from the transmitted PLC signal, the transformer already present for PLC communication can also be used for identifying an arc.

However, it is often necessary to keep the system comprising the transformer and the evaluation unit free of faults, for example to ensure that the arc identification and/or the PLC communication work properly.

Disclosure of Invention

The object of the invention is therefore to identify a fault in a system comprising a transformer for converting a primary signal of a direct voltage line into a secondary signal and an evaluation unit for evaluating the secondary signal.

According to the invention, this object is achieved by a method and a fault detection unit, wherein a signal component at least one frequency is determined from the secondary signal, the signal level of the signal component is compared with a signal threshold value, and a fault in the system is detected when the signal level is below the signal threshold value. In this way, the system, in particular the transformer, or parts of the evaluation unit can be checked for faults. For example, if the primary winding and/or the secondary winding of the transformer is damaged, the signal level of the signal component at the at least one determined frequency is below a signal threshold. If this is detected according to the invention, a fault in the system can be inferred. As a malfunction, an evaluation unit malfunction, for example, an amplifier failure, may also occur. Of course, different signal thresholds may be set for different frequencies.

The transformer can decouple the alternating primary signal from the dc voltage line via a primary winding connected in series with the dc voltage line and switch it to a secondary winding. Likewise, the transformer can also convert an alternating secondary signal into an alternating primary signal on the primary winding and couple it into the direct voltage line.

Advantageously, as the at least one frequency, at least one noise frequency of the fault-free system that typically occurs at the secondary winding is considered.

Thus, noise at the noise frequency can be considered as a signal component. A fault may be inferred if the signal level (noise level) at least one frequency (noise frequency) is below a signal threshold. Thus, by using the absence of noise in the secondary signal as an indication of a fault in the system, the noise can be positively utilized. For example, a typical noise source having a noise frequency may be the dc voltage bus.

Preferably, as signal threshold, the noise level of the fault-free system occurring at the noise frequency that typically occurs is used. However, the signal threshold may also be determined in a range greater than zero and less than the noise level that occurs at the noise frequency that typically occurs for a fault-free system.

Signal components at a plurality of frequencies of the frequency band can be found and signal levels at the plurality of frequencies are compared with signal thresholds, respectively. Thereby, the signal component comprises a plurality of frequencies at which the signal level is compared with a signal threshold (which may, but need not be, the same for the plurality of frequencies), respectively. When the signal level is below the signal threshold, a fault may be inferred.

Advantageously, signal components at a plurality of frequencies of the frequency band are determined, the signal levels at the plurality of frequencies are averaged and the average is compared with a signal threshold. By averaging, interference at frequencies in the frequency band can be ignored. Otherwise, despite the absence of a fault, this type of interference may cause the signal level at this frequency to be below the signal threshold. The total energy content in the frequency band is examined and evaluated by examining all signal levels of frequencies in the frequency band and averaging.

Preferably, the frequency band extends from 0Hz to a limit frequency of preferably 40 kHz.

Advantageously, the signal component is determined at system start-up. This ensures that the evaluation unit and the transformer operate properly and do not fail during startup. It is obvious that the signal component or the level of the signal component can also be determined during operation of the system.

The evaluation unit may represent itself by the fault identification unit or comprise a fault identification unit.

The evaluation unit can comprise an arc detection unit and/or a PLC receiving unit, which are designed to receive an arc signal and/or a PLC receiving signal converted from the primary winding to the secondary winding by means of a transformer.

The evaluation unit can also be part of an arc detection unit and/or a PLC receiving unit, which are designed to receive an arc signal and/or a PLC receiving signal converted from the primary winding to the secondary winding by means of a transformer. Thus, the arc detection unit and/or the PLC receiving unit may comprise an evaluation unit according to the invention.

A transmitting unit can also be provided, which is connected to the secondary winding of the transformer and is designed to transmit the PLC transmitting signal to the secondary winding. The transformer is then designed to transfer the PLC transmission signal from the secondary winding to the primary winding. The receiving unit on the dc voltage source or the dc voltage bus can retrieve and evaluate the modulated PLC transmission signal by demodulation.

The dc voltage line can be provided for transmitting a dc voltage of the at least one dc voltage source to the at least one dc voltage bus.

The at least one dc voltage source may comprise at least one photovoltaic cell and the at least one dc voltage bus comprises at least one inverter.

Preferably, a further signal component at least one further frequency is determined from the secondary signal and the signal level of the further signal component is compared with an upper signal threshold. When the signal level is above the upper signal threshold, a fault in the system is detected.

At least one further frequency may correspond to the at least one frequency, whereby the further signal component corresponds to the signal component.

Drawings

The invention is explained in detail below with reference to fig. 1 to 6, which show exemplary, schematic and non-limiting embodiments of the invention. Wherein:

figure 1a shows a system according to the invention when no fault has occurred,

figure 1b shows the system according to the invention in the event of a failure,

figure 2 shows a system suitable for receiving PLC received signals,

figure 3 shows a system suitable for detecting an arc,

figure 4 shows a system suitable for transmitting PLC receive signals and for detecting arcs,

figure 5 shows a system suitable for transmitting PLC receive signals and for detecting arcs when a short circuit occurs,

fig. 6 shows a typical frequency range of the secondary signal.

Detailed Description

In fig. 1a to 4, a DC voltage source 2 and a DC voltage bus 6 are shown, respectively, which are connected to one another via a positive DC voltage line DC + and a negative DC voltage line DC-. The dc voltage source 2 may comprise, for example, one or more solar cells and supplies energy, which is transmitted as a dc voltage U _ dc to the dc voltage bus 6. The direct voltage bus 6 may comprise, for example, an inverter, a rectifier, a step-up chopper and/or a step-down chopper, a DC-DC converter, a bidirectional converter or the like, and is used for feeding into an energy supply network or for the energy supply of a load. Of course, the electrical load can also be directly regarded as the dc voltage bus 6. In particular in photovoltaic installations, different configurations of the DC voltage source 2 and the DC voltage bus 6 can be provided, so that, for example, each solar panel is connected as a DC voltage source 2 via a DC voltage line DC +, DC-to an inverter as a DC voltage bus 6. The plurality of direct voltage sources 2 and/or the direct voltage bus 6 can also be divided into sections of direct voltage lines DC +, DC-.

Furthermore, the solar panels as direct voltage source 2 can also be connected in series and/or in parallel with each other. Various configurations of the DC voltage source 2, the DC voltage bus 6 and the DC voltage lines DC +, DC-are thus conceivable, whereby the invention is of course not restricted to the configuration shown in the figures. When a battery is used, the battery may be regarded as the dc voltage source 2 or the dc voltage bus part 6 depending on whether the battery is discharged or charged. Thus, depending on the operating mode, for example, the inverter or the charging device serves as the dc voltage source 2 in the case of a battery as the dc voltage bus 6, or the battery serves as the dc voltage bus 6 in the case of an inverter or a charging device as the dc voltage source 2.

Furthermore, a system 1 according to the invention is provided in fig. 1a to 4, the system 1 comprising a transformer 4, an evaluation unit 5 and a fault detection unit 7. The transformer 4 includes a primary winding L1 on the primary side and a secondary winding L2 on the secondary side. The primary winding L1 is connected in series with the negative DC voltage line DC-, and the secondary winding L2 is connected to the evaluation unit 5. Of course, the primary winding L1 can also be connected to the positive direct voltage line DC +.

In normal operation, the alternating primary signal i1, preferably the primary current, flowing through the primary winding L1 is converted by the transformer 4 into the alternating secondary signal i2, preferably the secondary current i2, flowing through the secondary winding L2 and vice versa. Advantageously, the transformer 4 has a conversion ratio of 1:1, 1:2 or 1:4 from the primary side to the secondary side. Furthermore, the communication transformer 4 may have a ferrite core, for example a Hiflux core, preferably with particularly advantageous saturation properties for direct currents.

The evaluation unit 5 is designed to evaluate the secondary signal i2 applied to the secondary winding L2. Thus, for example, a PLC receive signal rx (fx) and/or an arc signal arc (f _ arc) converted by the transformer 4 from the primary winding L1 to the secondary winding L2 may be received and processed, as will be described below with reference to fig. 2 and 3.

In the figure, it is exemplarily assumed that the transformer 4 has a conversion ratio of 1:1, whereby the primary signal i1 corresponds to the secondary signal i2 when the transformer 4 is functioning properly. Thus, when the transformer 4 is functioning properly, the primary signal i1 and the secondary signal i2 are substantially identical, whereby the frequency fn, the arc frequency f _ arc, the PLC-frequency fx, etc. on the primary and secondary sides are identical, however, only this case is assumed for a simpler description.

According to the invention, the secondary winding L2 of the transformer 4 is connected to the fault detection unit 7. The fault detection unit 7 may form part of the evaluation unit 5 or also be considered as evaluation unit 5 itself. The fault detection unit 7 determines a signal component n (fn) at least one frequency fn from the secondary signal i 2. The signal component i2(fn) has a signal level P (fn), which is compared with a signal threshold P _ min (fn). When the signal level P (fn) is below the signal threshold P _ min (fn), a fault F in the system 1 is detected. As a fault F, for example, damage occurring in the primary winding L1, the secondary winding L2 or also in the evaluation unit 5 or the fault detection unit 7 can be cited.

Advantageously, as signal component n (fn), the noise portion of the secondary signal i2 of the fault-free system 1, which typically occurs, is considered. Thus, the at least one frequency fn corresponds to the frequency of noise that typically occurs on the secondary winding L2. Accordingly, as signal threshold value P _ min (fn), the noise level that is predicted (since it typically occurs) in a fault-free system 1 at least one frequency fn (that is to say noise frequency) can be used. Likewise, the signal threshold P _ min (fn) may be determined in a range greater than zero and less than the noise level that occurs at the typically occurring noise frequency of the fault-free system 1.

In fig. 1a to 4, the signal components n (fn) are shown, for example, as arrows in the negative DC voltage line DC-. In the figure, the signal component n (fn) is thus switched from the primary winding L1 to the secondary winding L2. Of course, the signal component n (fn) can also be generated partially (for example at a certain frequency fn) or completely on the secondary side, for example on the secondary winding L2 or in the evaluation unit 5.

Now, if there is no fault F in the system 1, the signal component n (fn) at the at least one frequency fn has a signal level P (fn) above the signal threshold P _ min (fn), as shown in fig. 1 a. Thus, the failure F is not detected.

If a fault F occurs in the system 1 as shown in fig. 1b, the signal component n (fn) has a signal level P (fn) which is below the signal threshold P _ min (F2). In fig. 1a, the signal component n (fn) is no longer contained in the secondary signal i2, for example, because the line of the secondary winding L2 is interrupted (and thus the entire secondary signal i2 is interrupted in practice here). Thus, the presence of a fault F can be inferred in the fault detection unit 7.

The fault detection unit 7 according to the invention can be combined with different transformers 4. For example, as shown in fig. 2, the transformer 4 may be provided as a PLC transformer. Only then is a PLC receive signal rx (fx) having a PLC frequency fx converted from the primary winding L1 to the secondary winding L2 by the transformer 4 and received by, for example, a PLC receive unit which is part of the evaluation unit 5.

The transformer 4 can also be used to detect an arc, as shown in fig. 3. If an arc forms in the direct voltage source 2, the direct voltage busbar 6, the plug connection of the direct voltage lines DC +, DC-, or the like, an arc signal arc (f _ arc) with an arc frequency f _ arc is generated in the direct voltage lines DC +, DC-. The arc frequency f _ arc extends, for example, over a frequency spectrum of 5kHz to 200 kHz. Thus, the primary signal i1 includes an arc signal arc (f _ arc).

For example, in fig. 3, the electric arc is indicated as a lightning sign in the negative direct voltage line DC-. The transformer 4 converts the primary signal i1 into a secondary signal i2, whereby the secondary signal also comprises an arc signal arc (f _ arc). That is, here, the evaluation unit 5 comprises an arc detection unit and, in the event of an arc, can identify an arc signal arc (f _ arc) in the secondary signal i 2. From this, it can be concluded that an arc is present.

For a system 1 with an evaluation unit 5 comprising an arc detection unit and/or a PLC receiving unit, the identification of a fault F can be carried out similarly to that described in fig. 1a, b, in particular when no PLC receive signal Rx or no arc signal arc (F _ arc) is received at all.

However, if PLC receive signal rx (fx) is received, secondary signal i2 of course comprises PLC receive signal rx (fx). Reasonably, in order to not disturb the fault identification by the PLC frequency fx, the PLC reception signal rx (fx) has a PLC frequency fx different from the at least one frequency fn. This precondition is given, in particular, when a noise frequency is used as at least one frequency fn in order to avoid PLC communication in the frequency range of the noise. Furthermore, in the event of a fault situation (e.g. a line or winding interruption of the transformer 4), usually neither the PLC receive signal rx (fx) nor the signal component n (fn) is present in the secondary signal i2, as a result of which a fault F can be quickly inferred.

It is also advantageous for the arc detection that the arc does not interfere with the fault detection, i.e. the arc frequency f _ arc of the arc signal arc (f _ arc) is not in the range of the at least one frequency fn. However, even if the arcing signal arc (F _ arc) should be in the signal component n (fn), the identification of the fault F can be achieved when the signal level P (fn) of the signal component n (fn) at the time of the fault F is below the signal threshold value P _ min (fn). Thus, in particular when the winding or line of the transformer 4 is broken and damaged and thus the arc signal arc (f _ arc) is not applied to the evaluation unit 5, detection of the arc may not be possible. However, according to the invention, this fault is identified by the fault identification unit 7, which is also possible when the system is in operation.

A transmitting device 3 connected to the secondary winding L2 can also be provided for transmitting a PLC transmission signal tx (fx) having a PLC frequency fx, as is shown in fig. 4. The PLC transmit signal tx (fx) is modulated into a secondary signal i2 and converted into a primary signal i1 by transformer 4. The primary signal i1 is coupled into the direct-current voltage line DC +, DC-, and is transmitted via the direct-current voltage line DC +, DC-to the direct-current voltage source 2 and/or the direct-current voltage bus 6 and is received and demodulated by the receiving unit 20 provided on the direct-current voltage source 2 and/or the direct-current voltage bus 6. When the PLC transmission signal Tx is transmitted via the DC voltage line DC +, DC-, a filter capacitor Cf in or on the DC voltage bus 6 may also be used for passing the PLC transmission signal Tx through the circuit. For example, pulses for detecting interference sites, signals for measuring impedance, interference level measurement signals, synchronization signals for the individual current sources 2 (for example solar cells), or likewise control signals can be used as PLC transmission signals tx (fx).

The PLC frequency fx for the PLC receive signal rx (fx) and for the PLC transmit signal tx (fx) is typically in a frequency spectrum of 125kHz to 145kHz, for example 131.25kHz to 143.75 kHz.

Particularly advantageous is a system 1 having a fault detection unit and an evaluation unit 5 for detecting an arc, wherein additionally a transmission unit 3 for transmitting a PLC transmission signal tx (fx) is provided. Thereby, the transformer 4 can be used for PLC communication, but also for detecting an arc. It is possible to transmit the PLC transmission signal tx (fx) by the transmission means 3 and at the same time identify the arc by, for example, an arc detection unit which is part of the evaluation unit 5. Fig. 4 shows an advantageous embodiment of the invention, which makes it possible to identify an arc by means of an arc signal arc (f _ arc) and to transmit a PLC transmission signal tx (fx), wherein, in addition, a fault identification 7 is implemented. The primary winding L1 is connected to the direct voltage line DC +, DC-. The secondary winding L2 is connected to an evaluation unit 5, which in this case essentially forms an arc detection unit and comprises a fault detection unit 7. In order to attenuate the PLC transmit signal tx (fx) relative to the arcing signal arc (f _ arc), a resistor R and a series capacitor C of the order of 70 to 120nF are provided, for example, wherein the fundamental frequency can be set in the kHz range, preferably 130 kHz. A capacitor C is connected in parallel with the secondary winding L2. Furthermore, the evaluation unit 5 comprises a resistor R for converting the secondary signal i2 into a voltage U applied across the resistor R and for processing to identify an arc. Likewise, the secondary signal i2 is used by the voltage U to identify a fault F by examining the signal level P (fn) of the signal component n (fn) at the at least one frequency fn and comparing it with a signal threshold value P _ min (fn).

Since the transmitting unit 3 is likewise connected in series with the secondary winding, a resonant circuit is thus created, which, in the case of the evaluation unit 5 (together with the fault detection unit 7), leads to a PLC transmission signal tx (fx) which decreases up to 1/10 with respect to the signal component n (fn) and the arc signal arc (f _ arc). This enables an arc, for example a fault F, to be reliably detected, in particular even when transmitting the PLC transmission signal tx (fx). The embodiment shown in fig. 4 can of course also be implemented without arc detection. In other words, it is thus also possible to detect the transformer 4 with the transmitting unit 3 for PLC communication connected to the secondary winding L2 in terms of function or fault F without the presence of an arc detection unit. Of course, the signal component n (fn) can also be generated by the noise of the evaluation unit 5 itself. From this, a fault in the evaluation unit can also be inferred.

From the secondary signal i2, a further signal component n1(fn1) at least one further frequency fn1 is determined and the signal level P (fn1) of the further signal component n1(fn1) is compared with an upper signal threshold P _ max (fn 1). When the signal level P (fn1) exceeds the upper signal threshold P _ max (fn1), a fault F in the system 1 is detected. Fig. 5 shows a faulty system 1, which system 1 has the capacitance short KS of the evaluation unit 5 as a fault F. Here, the short-circuit KS likewise shorts the secondary winding L2. Thus, due to the short-circuit KS, the signal level P (fn1) of the further signal component n1(fn1) at the further frequency fn1 is above the upper signal threshold P _ max (fn1) and a fault F can be inferred, at which the short-circuit KS is present. Of course, the short-circuit KS of the capacitor C is described by way of example only, and other short-circuits KS or other faults in the evaluation unit 5 or in other components of the device 1 can also be identified.

In particular, the signal threshold P _ min (fn) and/or the on-signal threshold P _ max (fn1) may vary for different frequencies fn or for further frequencies fn1, when it is assumed that the fault F affects the levels P (fn), P (fn1) at different frequencies fn or further frequencies fn1 in a different manner. If the further frequency fn1 corresponds to the frequency fn, no fault is detected when the signal levels P (fn), P (fn1) are between the lower signal threshold P _ min (fn) and the upper signal threshold P _ max (fn 1). If the signal levels P (fn), P (fn1) are below the lower signal threshold P _ min (fn) or above the upper signal threshold P _ max (fn1), then a fault F is detected.

It may also be assumed that different types of faults F may reduce the signal levels P (fn), P (fn1) of different signal components n (fn) at different (further) frequencies fn, fn1 in the system 1. Thus, by comparing the levels P (fn), P (fn1) of the different (further) signal components n (fn), n1(fn1) at the different (further) frequencies fn, fn1 with the signal threshold P _ min (fn) and the on-signal threshold P _ max (fn1), it is possible to deduce the different fault types F in the system 1.

Fig. 6 shows a typical course of the secondary signal i2 in this frequency range. Here, as the signal component n (fn), for example, a frequency band of 0 to 40kHz is set as a cutoff frequency. The secondary signal i2 is shown in solid lines without a fault F. It can be seen that the signal component n (fn) is above the signal threshold P _ min (fn) for all frequencies fn, from which it can be concluded that no fault F has occurred in the system 1.

The secondary signal i2 is shown in dashed lines in the event of a fault F. The signal component n (fn) of the secondary signal i2 has a lower signal level p (fn) in the presence of a fault F compared to the signal component n (fn) without a fault F. Thus, at all frequencies fn, the signal level is below the signal threshold P _ min (fn). It can be concluded therefrom that no fault F has occurred in the system 1.

The signal level P (fn) at the frequency fn of the band can be averaged and compared to a signal threshold P _ min (fn) to find interference at a single frequency fn. Thus, instead of examining and evaluating the signal level p (fn) for a single frequency fn, the total energy content in the frequency band is examined and evaluated. Here, the average value may be an arithmetic average value, a median, or the like.

Of course, in addition to signal component n (fn), arc signal arc (f _ arc), PLC transmit signal tx (fx), and PLC receive signal rx (fx), primary signal i1 and secondary signal i2 may contain other parts of the alternating current, such as other signals, interference, etc.