Method and device for testing a transformer

Abstract

The present invention relates to a method (200) for testing a transformer (180). The method comprises determining (202) a flank pitch for a test signal (702) based on the transformer (180) to be tested, and generating (204) the test signal (702) with the determined flank pitch. The method also comprises testing (206) the transformer (180) using the generated test signal (702).

Description

[0001] Method and apparatus for testing a transformer

[0002] FIELD OF THE INVENTION

[0003] The present invention relates to a method and a correspondingly designed device for testing a transformer. In particular, the present invention is suitable for tests and measurements on transformers used as current transformers. Testing the transformer may, for example, include measurements of magnetization curves.

[0004] BACKGROUND OF THE INVENTION

[0005] In power engineering, instrument transformers, especially current transformers, are used. These transformers are based on a transformer designed for metrological requirements, enabling the potential-free measurement of large alternating currents under high electrical voltage. To ensure the proper condition of a current transformer, the current transformer must be tested for requirements defined, for example, in the EN 61869-2 standard. These requirements include, among other things, magnetization parameters, whereby a magnetization curve and saturation magnetization may be particularly relevant.

[0006] For this purpose, a test device can, for example, generate a sinusoidal signal at the mains frequency, which is fed into the device under test, such as a current transformer. This allows voltages and currents to be generated on the secondary side of the current transformer that also occur during nominal operation. For conventional current transformers, test signals with a power of up to 5 kW, for example, are therefore required. Due to this high output power, these test devices are either very heavy or can only test current transformers with low power ratings.

[0007] In addition, there are test procedures that use test signals with a reduced test frequency or test voltage to test the respective current transformer. However, these are generally very inaccurate because the changed behavior of current transformers when operating at a frequency deviating from the nominal or mains frequency is not sufficiently taken into account.

[0008] In this context, EP 1398644B1 discloses a method for testing a transformer, in which a test signal whose frequency is lower than the nominal or operating frequency of the transformer is applied to the transformer in order to measure a plurality of parameters of the transformer, in particular frequency-dependent parameters. A simulation model is derived from the parameters measured with this test signal, which simulation model enables the behavior of the transformer at different frequencies to be simulated. With the help of this simulation model or the parameters measured with the test signal, conclusions can then be drawn about the behavior of the transformer when operated at a frequency different from the frequency of the test signal, for example about the behavior of the transformer when operated at the nominal frequency. In particular, a square wave signal whose fundamental frequency is lower than the nominal frequency can be used as the test signal.

[0009] The use of square-wave signals is particularly suitable for achieving a sufficiently large magnetic flux in the magnetic core. However, this can lead to the problem that capacitive current peaks can occur at high edge steepnesses of the square-wave signal in combination with high voltages in converters with a design-related high parasitic capacitance. These currents can overload the test device or require correspondingly larger and thus more expensive and heavier test devices. Furthermore, this can impair the precision of the measurement.

[0010] SUMMARY OF THE INVENTION

[0011] An object of the present invention is therefore to provide a precise measurement of magnetization characteristics of a transformer, in particular a magnetization curve of a current transformer, which can be realized cost-effectively and is suitable for a wide range of instrument transformers.

[0012] This object is achieved according to the invention by a method for testing a transformer and a correspondingly designed device for testing a transformer, as defined in the independent claims. The dependent claims define embodiments of the present invention.

[0013] According to the invention, in a method for testing a transformer, an edge steepness for a test signal is determined as a function of the transformer to be tested, and the test signal is generated with the edge steepness determined in this way. The transformer is tested using the test signal generated. In other words, a test signal is used with a signal shape which has an edge steepness that depends on the transformer to be tested. The signal shape of the test signal used therefore has defined harmonics, i.e. defined components of frequencies above a fundamental frequency of the test signal. In particular, this can ensure that the signal shape of the test signal has fewer high frequency components than, for example, a square-wave signal, as is usually generated by means of constant voltage sources connected in a bridge.Furthermore, undershoots and overshoots can be avoided or reduced by adjusting the edge steepness. The edge steepness of the test signal is not determined by a switching mechanism which is used in the test signal generation, but is set as a function of the transformer. An appropriately controllable signal source can be used for this purpose. By adjusting the edge steepness, in particular a lower edge steepness compared to connected constant voltage sources, large capacitive current peaks can be avoided and thus on the one hand overloading of the device which generates the test signal can be avoided and on the other hand measurement accuracy can be increased because the currents to be measured occur in a limited range which can therefore be recorded precisely.By determining and adjusting the edge steepness depending on the transformer, current transformers with a high capacity due to their design can also be tested, and in particular a magnetization curve can be measured precisely.

[0014] The test signal can be generated, for example, by controlling at least one adjustable voltage converter. The voltage converter can be a direct current converter, a so-called DC-DC converter. In order to generate an alternating voltage with the desired edge steepness, for example, two voltage converters can be used. One of the two voltage converters generates an adjustable positive voltage referenced to ground, and the other of the two voltage converters generates an adjustable negative voltage referenced to ground. The adjustable positive voltage and the adjustable negative voltage can each be set to a variety of voltage values ​​by a control device and are fed to the transformer to be tested. The previously determined edge steepnesses for rising and falling edges can be realized by appropriately adjusting the voltage over time.An adjustable range of the output voltage of the voltage converters can be, for example, zero to 25 V or zero to 50 V, for example in steps of 1 V or 0.1 V.

[0015] By adjusting the positive and negative voltages relative to ground, an iron core of the transformer can be magnetized in both directions, for example to measure magnetization curves.

[0016] To determine the edge steepness for the test signal, for example, a parasitic capacitance of the transformer can be determined . The edge steepness can be determined as a function of the parasitic capacitance and a specified maximum current that may occur during testing of the transformer . Parasitic capacitances can occur, for example, in connecting cables, connecting terminals and the design of the transformer . These capacitances can lead to current peaks, particularly at high frequencies . By setting the edge steepness for the test signal, high frequencies can be avoided, for example by reducing the edge steepness . This can prevent current peaks . The test device is subjected to less strain because it has to provide a lower maximum current . This can improve the reliability of the test device and make the test device more compact and lighter.

[0017] The parasitic capacitance can be determined by a measurement on the transformer under test. In particular, a high-frequency measurement can be carried out on the transformer, for example. A high-frequency measurement on the transformer means, for example, that a frequency at which the parasitic capacitance is measured is higher than the nominal frequency, for example a multiple of the nominal frequency. The nominal frequency of the transformer can be, for example, 50 or 60 Hz. The high frequency used in measuring the parasitic capacitance can be, for example, more than 1 kHz or even more than 100 kHz. The voltage used for this purpose can be correspondingly small, in particular smaller than the voltage to be used when testing the transformer. For example, a test of the transformer can be carried out at + / - 25 V, whereas the parasitic capacitance is measured at + / - 5 V.

[0018] Alternatively or additionally, the parasitic capacitance can be entered by a user. The user may have determined the value of the parasitic capacitance, for example, using an additional measuring device, or the parasitic capacitance may be stored in a database. Furthermore, the type of transformer to be tested and, if applicable, its connection configurations can be entered, whereupon the parasitic capacitance is retrieved from, for example, a database.

[0019] It is clear that a currently measured parasitic capacitance is advantageous because in this case the current conditions of the transformer are taken into account and the edge steepness for the test signal can be optimally set. An appropriately designed test device can, for example, first automatically determine the parasitic capacitance by measurement and then begin the actual test of the transformer, e.g. a measurement of a magnetization curve of the transformer, using the edge steepness determined from the parasitic capacitance. Taking the parasitic capacitance into account can increase the measurement accuracy. Firstly, current peaks due to the parasitic capacitance can be reduced and determined by the selected edge steepness, so that a distinction can be made between regular inductive current through the transformer and capacitive current peaks.For example, a magnetization curve of the transformer can then be measured with great precision using the regular inductive current thus determined.

[0020] According to one embodiment, the test signal is a periodic test signal. The edge steepness can be determined using an iterative method. In the iterative method, based on an edge steepness in one period of the test signal, an edge steepness for a following period of the test signal is determined as a function of a measured maximum current that

[0021] Testing the transformer with the test signal generated in the previous period. For example, the slew rate can be increased slightly from one period to the next. For example, starting from a relatively low slew rate, the slew rate can be increased slowly, thereby increasing the capacitive current peaks. This can be continued until a desired or permissible maximum current is reached. The maximum current can, for example, correspond to a rated current of the test device or a certain percentage of the maximum current of the test device, for example 90% of the maximum current. Once the slew rate has been determined using this iterative procedure, the measurement on the transformer can be carried out using the determined slew rate.

[0022] The test signal can, for example, be a periodic voltage signal. For each period, the voltage signal can comprise a voltage rise with the specific edge steepness from a lower voltage value to an upper voltage value, an upper section with a substantially constant voltage at the upper voltage value, and a voltage drop from the upper voltage value to the lower voltage value. It is clear that the lower voltage value is smaller than the upper voltage value. The lower voltage value can, for example, be negative with respect to ground and the upper voltage value can be positive with respect to ground. The absolute values ​​of the upper voltage value and the lower voltage value can be different or the same. For each period, the voltage signal can have a lower section with a substantially constant voltage at the lower voltage value.A length of the lower section and a length of the upper section may be substantially equal. The voltage drop may have a slope equal in magnitude to the

[0023] The slope of the voltage rise is .

[0024] It is clear that the lower section may also have a different length than the upper section. Likewise, the voltage drop may have a different slope than the voltage rise. In many embodiments, however, the upper section will have substantially the same length as the lower section, and the slopes of the voltage drop and the voltage rise will have substantially the same slope.

[0025] In these cases the voltage signal is symmetrical to ground.

[0026] The slope can be constant over the rise time or fall time, so that the test signal has a trapezoidal signal waveform. However, the slope can also vary over the rise time or fall time, for example, starting with a small slope, which then increases and reaches its maximum near the zero crossing of the voltage signal, before decreasing again. This results in a trapezoidal signal waveform with "smooth" transitions between the essentially constant plateau sections and the slopes.

[0027] According to further embodiments, a point in time at which an edge of the test signal begins is set or initiated depending on a comparison of a current through the transformer generated by the test signal with a threshold value. As described above, the test signal can be periodic with a substantially constant upper section and a substantially constant lower section. An edge of the test signal begins when, for example, a voltage drops from the substantially upper (positive) voltage value or a voltage rises from the substantially lower voltage value. During a test of the transformer, for example, the substantially constant voltage in the upper section causes a (positive) current through a winding of the transformer to rise due to the inductance of this winding.The current through the winding can be measured and, when the current reaches the threshold value, a voltage drop with the previously determined edge steepness can be initiated. However, as long as the voltage is still positive during the falling edge, the current through the winding continues to rise. Only after the voltage crosses zero is the current through the winding reduced. The voltage reaches the lower (negative) voltage value and then remains essentially constant. The (positive) current is further reduced, becomes zero and then flows in the opposite direction (negative current) through the winding and increases in magnitude, while the voltage value remains essentially constant at the lower voltage value. As soon as the (negative) current through the winding has reached the threshold value, the rising edge begins.As long as the voltage is still negative during the rising edge, the (negative) current continues to rise. Only when the rising edge reaches zero and a positive voltage is then applied to the transformer again does the (negative) current decrease. While the voltage has the essentially constant upper (positive) voltage value, the (negative) current initially decreases to zero and then increases again in the opposite direction (positive current) until it reaches the threshold value. Then the previously described process begins.

[0028] Process again .

[0029] As previously described, the current initially continues to rise with the onset of the edges because the edges do not have an infinite steepness and the voltage therefore remains positive (with a falling edge) or negative (with a rising edge) until the zero crossing. The steepness of the edges is rather small, so that high frequency components are not included in order to avoid current peaks caused by parasitic capacitances. The further increase in current even after the onset of the edges is therefore unavoidable and must be taken into account in order not to exceed a desired maximum current. For example, to avoid overloading the test device.The magnitude of the threshold value at which the edges are initiated can therefore be set to an amount of a maximum rated current of the test device which provides the test signal, for example to a value in the range of 70 to 95% of the rated current, for example to 90% of the rated current, or depending on an expected further increase in the inductive current, which in turn depends on the set edge steepness.

[0030] For example, the magnitude of the threshold value can be set dynamically as a function of an observation of the current through the transformer. For example, the threshold value can be determined as a function of the edge steepness of the test signal, a voltage swing of the test signal and a magnitude increase in the current through the transformer generated by the test signal. The voltage swing of the test signal is, for example, the voltage difference between the upper and lower voltage values ​​in a periodic test signal. Assuming, for example, that the edge steepness is constant, i.e. that the voltage drop per unit of time is constant from the beginning to the end of the edge, the time from the beginning of the edge to the voltage zero crossing can be determined in a simple manner.Even if the edge does not have a constant slope, since the edge shape is determined and adjusted, the time from the start of the edge to the voltage zero crossing can be determined. During this time, as discussed previously, the current through the transformer continues to rise, at least due to the inductance of the transformer winding. In addition, the current may increase due to parasitic capacitances. The current rise of the current flowing through the transformer can be measured, for example by measuring the current continuously during test of the transformer and determining the corresponding current change, that is, change in current per unit time. Continuous can mean, for example, that the current is measured continuously at short time intervals of, say, 1 ps and the current change is determined from the change in current per microsecond.The maximum period over which the current can continue to rise is determined from the time from the start of the edge to the voltage zero crossing. An estimate of the further voltage rise that can be achieved in this time can be made, for example, by multiplying the last determined current rise by the time from the start of the edge to the voltage zero crossing. Based on this estimate, the threshold can be set so that the current does not exceed a desired maximum value, for example the maximum rated current of the test device. In other words, the threshold can be determined as the difference between an amount of a desired maximum current generated by the test signal through the transformer and a product of the current current rise and a polarity reversal time of the test signal. The polarity reversal time depends on the voltage swing of the test signal and the edge steepness.

[0031] An additional reserve of, for example, 10% of the maximum current can be provided for additional current due to parasitic capacitances. However, this reserve can be kept relatively small, since the specifically adjusted edge steepnesses ensure that the additional current due to parasitic capacitances is relatively small compared to corresponding current peaks due to edge steepnesses obtained by simply switching the test signal between the upper and lower voltage values.

[0032] According to one embodiment, the testing may comprise determining a magnetization curve of the transformer as a function of the periodic voltage signal and a time profile of a current generated by the voltage signal through the transformer.

[0033] According to the present invention, a device for testing a transformer is further provided, that is to say a test device. The device comprises a test signal source for generating a test signal for the transformer and a processing device which is coupled to the test signal source. The processing device is designed to determine an edge steepness for the test signal as a function of the transformer to be tested, to generate the test signal with the edge steepness thus determined by means of the test signal source, and to test the transformer using the generated test signal. The device is thus suitable for carrying out the method described above and therefore also comprises the advantages described above.

[0034] The features of the previously described embodiments and aspects of the invention can be combined with one another in any way, unless expressly stated otherwise. In particular, the features can be used not only in the described combinations, but also in other combinations or on their own.

[0035] SHORT DESCRIPTION OF THE CHARACTERS

[0036] The present invention will be described in detail below with reference to the accompanying figures.

[0037] Fig. 1 shows schematically a device for testing a transformer according to an embodiment of the present invention in connection with a transformer.

[0038] Fig. 2 shows schematically method steps of a method for testing a transformer according to an embodiment of the present invention.

[0039] Fig. 3 shows schematically a test signal source according to an embodiment of the present invention

[0040] Fig. 4 shows schematically a test signal source according to a further embodiment of the present invention.

[0041] Fig. 5 schematically shows a test signal source according to yet another embodiment of the present invention. Fig. 6 schematically shows a voltage curve of a test signal fed into a transformer and an associated current curve.

[0042] Fig. 7 shows schematically a voltage curve of a test signal which is fed into a transformer, with a slope which is determined according to a

[0043] embodiment of the present invention, and an associated current profile.

[0044] Fig. 8 schematically shows method steps for determining a slope according to an embodiment of the present invention.

[0045] Fig . 9 shows schematically a determination of a threshold value from a current waveform which is used to determine a time at which an edge of the test signal begins .

[0046] DETAILED DESCRIPTION OF EMBODIMENTS

[0047] The above-described properties, features and advantages of this invention, as well as the manner in which they are achieved, will become clearer and more readily understood in connection with the following description of embodiments which are explained in more detail in connection with the drawings.

[0048] The present invention will be described below with reference to

[0049] Embodiments are explained in more detail with reference to the drawings. In the figures, like reference numerals designate like or similar elements. The figures are schematic representations of various embodiments of the invention. Elements shown in the figures are not necessarily drawn to scale. The various elements shown in the figures are shown in such a way that their function and general purpose will be understood by those skilled in the art. Connections and couplings between functional units and elements shown in the figures can also be implemented as an indirect connection or coupling. Functional units can be implemented as hardware, software or a combination of hardware and software.

[0050] Fig. 1 shows a transformer 180 in conjunction with a device 100 for testing the transformer 180. The transformer 180 can, for example, be a current transformer provided in a power engineering electrical system. The device 100 can, for example, be coupled to a secondary side of the transformer 180. The device 100 can, for example, comprise a portable test device which is transported to the transformer 180 for testing the transformer 180 and is coupled to the transformer 180 via lines 150, 152. In this context, the transformer to be tested is also referred to as a device under test (DUT).

[0051] The device 100 comprises a test signal source 102 and a processing device 104. The processing device 104 can, for example, comprise a microprocessor controller which comprises a working memory, a mass storage device for program code and data, and input / output devices. Furthermore, the device 100 can comprise a user interface 106 and a current measuring device 108. The user interface 106 can, for example, have a possibly touch-sensitive screen, a keyboard or keys, a loudspeaker, signal lights, and the like. The current measuring device 108 can, for example, be provided in or on a line to the transformer 180, e.g., in the line 152, in order to measure a current flowing through the transformer 180, which current flows due to a voltage generated by the test signal source 102.The current measuring device 108 can comprise an analog / digital converter which determines the current strength at a predetermined sampling frequency and provides it as a corresponding digital value to the processing device 104. The test signal source 102, the user interface 106 and the current measuring device 108 are coupled to the processing device 104 and are controlled by it and / or supply measured values ​​and other information to the processing device 104, for example user inputs. The device 100 can comprise further components, for example a power supply device for supplying the previously described components with electrical energy and further measuring devices for measuring current, voltage or power on the lines 150, 152. The device 100, in particular the processing device 104, can be configured to carry out the methods and techniques described below.The device 100 can also be designed to carry out further tests on other test objects, for example high-voltage switches, electrical feedthroughs or electrical lines.

[0052] When testing a transformer, it may be necessary to feed test signals into the transformer which do not correspond to the nominal signals. These test signals can, for example, have lower or higher frequencies than the nominal frequency of the transformer. This makes it possible to test properties of the transformer which cannot be tested at the nominal frequency or can only be tested with great effort, for example saturation magnetization. Possible test signals are sinusoidal signals and, in particular for determining saturation magnetization, also rectangular signals, which, however, also contain very high frequency components. A test signal can be a voltage signal, for example. In response, a current flows whose temporal characteristic, i.e. the current strength as a function of time, can be recorded by the current measuring device 108. With the help of a model, the corresponding nominal values ​​can be calculated back from the results of the test signals.The use of square-wave signals is particularly necessary for achieving a sufficiently large magnetic flux in the magnetic core. However, the high edge steepness of the square-wave signal, in conjunction with design-related parasitic capacitances, leads to large capacitive current peaks. The parasitic capacitances can arise, for example, from the connections and connecting leads as well as the design of the transformer itself. The large current peaks can overload the test signal source 102 and / or impair the accuracy of the measurement.

[0053] Therefore, the test signal source 102 generates a test signal in which the high frequency components are reduced compared to a switched square wave signal, i.e. a test signal with defined rising and falling edges. For example, instead of a simple square wave signal, a combination of triangular and square wave signals can be generated, i.e. a test signal with a trapezoidal shape. In order to still achieve the longest possible

[0054] In order to have sections with constant voltage in the test signal, for example to achieve saturation magnetization, and additionally to be able to provide high frequency components in the test signal, which can be used, for example, by the model to determine properties of the transformer, a method 200 is carried out, for example under the control of the processing device 104, as shown in Fig. 2.

[0055] In method 200, in step 202, an edge steepness for the test signal is determined as a function of the transformer 180 to be tested, and in step 204, the test signal with the determined edge steepness is generated by means of the test signal source 102. The transformer 180 is then tested in step 206 using the generated test signal. The goal is to determine the edge steepness for the test signal such that long sections with constant voltage and high frequency components are included in the test signal, without, however, overloading the test signal source 102 with capacitive current peaks and impairing the accuracy of the measurement.

[0056] Technically, the test signal with adjustable edge steepness can be realized by using one or more voltage sources that have an adjustable output voltage. In a conventional generation of a square wave, for example, two DC voltage sources with a fixed output voltage can be used. One of the DC voltage sources has, for example, a positive output voltage referenced to ground and the other DC voltage source has a negative output voltage referenced to ground. The two output voltages can be equal in magnitude. For example, via a

[0057] Using a changeover switch or two switches, either the DC voltage source with the positive output voltage or the DC voltage source with the negative output voltage is connected to one terminal of a transformer winding, while the other terminal of the transformer winding is connected to ground. However, it is not possible to adjust the edge steepness.

[0058] The test signal with the adjustable edge steepness can be generated, for example, by using and controlling at least one adjustable voltage converter. The voltage converter can be, for example, a direct current voltage converter, a so-called DC / DC converter. The adjustable voltage converter can, for example, be selectively adjustable to one of a plurality of voltage values ​​under the control of the processing device 104, for example to a voltage between zero volts and 25 V or 50 V with an accuracy of 1 V or 0.1 V. For example, the voltage converter can comprise a clocked or linear direct current amplifier (DC-DC amplifier) ​​that can generate positive and / or negative output voltages. The voltage converter or amplifier is capable of providing the necessary voltages and currents with the necessary dynamics.

[0059] Fig. 3 schematically shows an example of an implementation of the test signal source 102. In this example, the test signal source 102 comprises two adjustable voltage converters 302, 304 and two switches 306, 308. The switches 306, 308 can be mechanical or electronic switches controllable by means of the control device 104. The voltage converter 302 is designed, for example, to output an output voltage Ui, which is negative with respect to a ground line 310 and can be set to a plurality of values ​​between zero and a maximum voltage U max is adjustable. U max can, for example, be in a range from 10 V to 100 V. For example, U max 25 V. The voltage converter 304 is designed, for example, to provide an output voltage U2 which is positive relative to the ground line 310 and whose magnitude can be set to a plurality of values ​​between zero and U maxis adjustable. The output voltages of the voltage converters 302, 304 can be adjusted by the control device 104. Via the switches 306, 308, either the output voltage of the voltage converter 302 or the output voltage of the voltage converter 304 can be connected to the line 312. By appropriately controlling the voltage converters 302, 304 and the switches 306, 308, an alternating voltage in the range of -U max to +U maxon the lines 310, 312, which can be used to test the transformer 180, as shown in Fig. 1. For example, a periodic trapezoidal alternating voltage can be provided. A fundamental frequency of this periodic alternating voltage can, for example, correspond to a nominal frequency of the transformer 180, for example 50 Hz or 60 Hz. It is clear that the fundamental frequency can have any other value above or below the nominal frequency of the transformer, for example 30 Hz or 100 Hz. Within a period, the alternating voltage can, for example, have the following voltage characteristic: (1) a voltage rise with a desired edge steepness from a lower voltage value, for example -U max , to an upper voltage value, for example +U max; (2) an upper section with essentially constant voltage at the upper voltage value; (3) a voltage drop from the upper voltage value to the lower voltage value with a desired slew rate; and (4) a lower section with essentially constant voltage at the lower voltage value. It is clear that the lower voltage value and the upper voltage value can be different in magnitude. It is also clear that the slew rate from the lower voltage value to the upper voltage value can be different from the slew rate from the upper voltage value to the lower voltage value. In many examples, however, the lower voltage value is equal in magnitude to the upper voltage value and the slew rate of the voltage increase is equal in magnitude to the slew rate of the voltage decrease.For a voltage swing between the upper and lower voltage values ​​of, for example, 50 V, the desired edge steepness can be, for example, 50 V per 28 ps. One period of an exemplary 50 Hz test signal is 20 ms, so the lower and upper sections are each just under 10 ms long (exactly 9.972 ms).

[0060] Fig. 4 schematically shows a further example of an implementation of the test signal source 102. In this example, the test signal source comprises two adjustable voltage converters 402, 404. The adjustable voltage converter 402 is designed to provide an adjustable output voltage Ui on the line 408, which is positive with respect to a connecting line 406 connecting the two adjustable voltage converters 402, 404 and whose magnitude can be adjusted to a plurality of values ​​between zero and Umax. The adjustable voltage converter 404 is designed to provide an adjustable output voltage U2 on the line 410, which is also positive with respect to the connecting line 406 and whose magnitude can be adjusted to a plurality of values ​​between zero and U maxis adjustable. By suitable control of the voltage transformers 402, 404 by means of the control device 104, an alternating voltage with a voltage swing of 2xU max and adjustable edge steepness.

[0061] Fig. 5 schematically shows yet another example of an implementation of the test signal source 102. In this example, the test signal source comprises an adjustable voltage converter 502 and a fixed voltage source 504. The adjustable voltage converter 502 is designed to provide, by means of the control device 104, an adjustable output voltage Ui on the line 508, which is positive with respect to a connecting line 506, which connects the voltage converter 502 to the fixed voltage source 504, and whose magnitude can be set to a plurality of values ​​between zero and 2xU maxis adjustable. The fixed voltage source 504 is designed to provide a fixed output voltage U2 on the line 510, which is also positive with respect to the connecting line 506 and has, for example, the value U max By appropriately controlling the voltage converter 502, an alternating voltage with a voltage swing of 2xU max and adjustable edge steepness.

[0062] 6 and 7 show schematically and by way of example signal waveforms 602, 702 of an output voltage of the test signal source 102, as can be generated by suitable control of the previously described voltage converters in the test signal source 102. In Figs. 6 and 7, the signal waveform 602, 702 of the output voltage is shown in the upper part in each case and a respective signal waveform 604, 704 of a current through the transformer 180 connected to the test signal source 102 is shown in the lower part. The signal waveform 604, 704 of the current through the transformer 180 can be measured, for example, by means of the current measuring device 108 (see Fig. 1). The output voltage can, for example, have a voltage swing of 50 V, that is to say the upper voltage level is, for example, +25 V and the lower voltage level is, for example, -25 V. In the example, the ohmic resistance of the transformer 180 is 220 mQ and the inductance is 0.5 H.The frequency of the output voltage is essentially 50 Hz. Without taking into account the capacitive current peaks described below, the inductive current between the signal edges, i.e., during the upper and lower signal sections with essentially constant voltage, increases in magnitude to approximately 250 mA, for example.

[0063] The signal curve 602 of the output voltage shown in Fig. 6 has very steep signal edges. The voltage swing of the voltage shown in Fig. 6 can be 50 V, for example, and the edge steepness can be 25 V / ps or more, for example 50 V / ps. Due to the parasitic capacitances, which can be generated, for example, by lines and connections of the transformer or by the design of the transformer itself, very high currents 606, so-called current peaks, occur during the signal edge, as shown in the lower part of Fig. 6. The cause of these high currents are high-frequency components of the steep signal edges, which enable high capacitive currents. The amount of these current peaks can be considerably higher than the amount of the inductive current, for example 500 mA, 1000 mA or more.

[0064] The signal curve 702 of the output voltage shown in Fig. 7 has less steep signal edges. With a voltage swing of, for example, 50 V, the edge steepness can be, for example, 50 V / 20 ps or less, for example 50 V / 28 ps. As can be seen from the signal curve 704 of the corresponding current shown in Fig. 7, with this lower edge steepness, significantly lower current peaks 706 occur because the signal edges now have significantly lower amplitudes in the high frequency components. The capacitive current peaks 706 become correspondingly lower, for example only 50 mA or 100 mA.

[0065] A stream I c a capacitive current peak depends on the parasitic capacitance C of the transformer 180 as follows : ( Equation 1 )

[0066] In practice, values ​​for the parasitic capacitance often range between 100 pF and 30 nF, but are not limited to this range and can therefore be larger or smaller, for example, 3 pF. Therefore, the slope to be used is determined and adjusted depending on the transformer being tested. Various procedures are described below.

[0067] In one procedure for determining the edge steepness, the parasitic capacitance of the transformer is first determined. For this purpose, the parasitic capacitance can be measured on the transformer, for example, using a high-frequency measurement. The high-frequency measurement can, for example, be carried out at a lower voltage than the later test voltage in order to ensure that the device 100 is not overloaded. The parasitic capacitance can also be determined via an input from a user via the user interface 106. For example, the user can enter the parasitic capacitance directly from information about the transformer, or the user can enter a type of transformer and the parasitic capacitance can be determined, for example, from a database based on the type of transformer.The edge steepness can then be determined as a function of the parasitic capacitance and a predetermined maximum capacitive current that should not be exceeded during transformer testing. The maximum capacitive current can be determined, for example, from a maximum rated current of the test signal source 102 and a maximum inductive current occurring during transformer measurement, for example as a difference between these two currents. The edge steepness du / dt can then be determined from Equation 1 above.

[0068] A further procedure for determining the edge steepness is described in connection with a method 800 shown in Fig. 8. First, in step 802, an edge steepness is set to an initial low steepness. The initial low steepness can, for example, be so low that it does not lead to an overload of the test signal source 102 and does not impair the measurement results even with transformers having a very high parasitic capacitance (for example 3 pF). If, for example, the capacitive current is to remain below 100 mA, the initial low steepness can be set to 50 V / 1500 ps according to equation (1), for example, so that the test signal source is not overloaded even with a parasitic capacitance of 3 pF. A preliminary test signal is generated using this edge steepness (step 804), and the maximum occurring current is determined in step 806.If the maximum current occurring is less than the permissible maximum current, for example less than the nominal current of the test signal source 102 (step 808), the provisional edge steepness can be increased in step 810, for example by a certain value or percentage. The method is repeatedly continued with the new provisional edge steepness in step 804. If the permissible maximum current is reached in step 808, the edge steepness to be used for testing the transformer 180 can be determined in step 812; for example, the edge steepness at which the permissible maximum current has not yet been reached can be used.

[0069] As described above, the test signal can be a periodic signal. A frequency can be predetermined, for example, by test conditions; for example, the test signal can have the nominal frequency of the transformer or a multiple thereof as the fundamental frequency. The fundamental frequency of the test signal can also be lower than the nominal frequency of the transformer. However, particularly when measuring the magnetization curve of the transformer, it can be necessary to feed in the highest possible current in order to generate the highest possible magnetic flux in the transformer 180. As soon as this high current is reached, the polarity of the applied voltage signal must be switched in order to generate the highest possible current and thus the highest possible magnetic flux in the transformer in the opposite direction. This process can be repeated several times in succession when recording the magnetization curve.For this purpose, a comparator can be provided which, for example, in software in the processing device 104, compares the current from the current measuring device 108 with a threshold current, which corresponds, for example, to the nominal current of the test signal source 102. However, the reduced edge steepness delays the polarity reversal, since a certain time elapses during the falling or rising edge until the voltage reaches the zero crossing and reduces the inductive current flow and thus the magnetic flux, whereby the current can exceed the threshold current. To avoid this, reaching the threshold current can be predicted as follows.

[0070] Fig. 9 shows, by way of example, a non-linear current waveform 902 over time, which flows as an inductive current through a winding of the transformer 180 due to the voltage applied to the transformer 180. The current waveform 902 is non-linear because it generally involves saturating inductances. The inductance value is therefore unknown and current-dependent. Thus, the gradient of the current waveform 902 is also unknown. Therefore, a continuous determination of the gradient of the current waveform 902 is carried out. For example, the current can be sampled at successive points in time (for example at the vertical dashed lines shown in Fig. 9) and in this way an Ai / At of the current waveform can be continuously determined in real time. A delay td between the start of an edge and the zero crossing can be determined due to the edge steepness from the voltage swing and the set edge steepness.From the desired maximum current I. t hr can be a corrected maximum current I t hr korr can be determined from the current gradient Ai / At as follows:

[0071] ( Equation 2 ) thr_korr =I th r~^td

[0072] As soon as the measured current exceeds the corrected maximum current

[0073] Ithr korr is reached , the polarity reversal process is initiated with the determined edge steepness so that the maximum current I t hr is not exceeded.

[0074] Of course, the features of the previously described embodiments and aspects of the invention can be combined with one another. In particular, the features can be used not only in the described combinations, but also in other combinations or on their own, without departing from the scope of the invention.

Claims

CLAIMS 1. A method for testing a transformer, comprising: Determining (202) an edge steepness for a test signal (702) depending on the transformer (180) to be tested, Generating (204) the test signal (702) with the determined edge steepness, and Testing (206) the transformer (180) using the generated test signal (702).

2. The method of claim 1, wherein determining the slope comprises: Determining a parasitic capacitance of the transformer (180), Determining the edge steepness as a function of the parasitic capacitance and a predetermined maximum capacitive current that may occur during testing of the transformer (180).

3. The method of claim 2, wherein determining the parasitic capacitance comprises: Measuring the parasitic capacitance by means of a high-frequency measurement on the transformer (180), and / or Capturing a parasitic capacitance input from a user.

4. Method according to one of the preceding claims, wherein the test signal (702) is a periodic test signal, wherein the determination of the edge steepness comprises an iterative method (800) in which, based on an edge steepness in a period of the test signal, a following Edge steepness for a following period of the test signal (702) is determined as a function of a measured maximum current generated when testing the transformer (180) with the test signal.

5. The method according to any one of the preceding claims, wherein a point in time at which an edge of the test signal (702) begins is set as a function of a comparison of a current generated by the test signal (702) through the transformer (180) with a threshold value (Ithr kon).

6. The method according to claim 5, wherein the magnitude of the threshold value (Ithr kon) is less than an amount of a maximum rated current of a device (100) providing the test signal (702).

7. The method according to claim 5 or claim 6, wherein the test signal (702) comprises a voltage signal and the threshold value (Ithr ko ) is determined as a function of the edge steepness of the test signal (702), a voltage swing of the test signal (702) and an absolute current increase (Ai / At) of a current generated by the test signal (702) through the transformer (180).

8. The method of claim 7, wherein the current increase (Ai / At) of the current generated by the test signal (702) is determined continuously during testing of the transformer (180).

9. The method according to claim 8, wherein the threshold value (Ithr corr) is defined as the difference between an amount of a desired maximum current (Ithr) generated by the test signal through the transformer (180) and a product of the current increase (Ai / At) and a polarity reversal time (td) of the test signal (702) is determined, which depends on the voltage swing of the test signal and the edge steepness.

10. Method according to one of the preceding claims, wherein the test signal (702) is a periodic voltage signal.

11. The method of claim 10, wherein the periodic voltage signal comprises, per period: a voltage rise with the determined edge steepness from a lower voltage value to an upper voltage value, an upper portion with a substantially constant voltage at the upper voltage value, a voltage drop from the upper voltage value to the lower voltage value, and a lower portion with a substantially constant voltage at the lower voltage value.

12. The method according to claim 11, wherein the voltage drop has a slope which is equal in magnitude to the slope of the voltage rise.

13. The method according to any one of the preceding claims, wherein testing (206) of the transformer (180) comprises determining a magnetization curve of the transformer (180) as a function of the test signal (702) and a time profile (704) of a current generated by the test signal through the transformer (180).

14. Method according to one of the preceding claims, wherein generating (204) the test signal (702) comprises driving of at least one adjustable voltage converter (302, 304, 402, 404, 502).

15. Apparatus for testing a transformer, comprising: a test signal source (102) for generating (204) a test signal (702) for the transformer (180), a processing device (104) which is coupled to the test signal source (102) and is designed to determine (202) an edge steepness for the test signal (702) as a function of the transformer (180) to be tested, to generate (204) the test signal (702) with the determined edge steepness by means of the test signal source (102), and to test (206) the transformer (180) using the generated test signal (702).

16. Device according to claim 15, wherein the device (100) for carrying out the method (200) according to one of the Claims 1-14 are designed.

Images