The phenomena that take place in a transformer when it is turned on or when it is turned off are phenomena of an unsteady state, lasting only a fraction of a second. Despite the insignificant duration of these phenomena, their study is extremely necessary, since their consequences, if you do not take certain countermeasures, can damage the transformer or the devices included in its circuit. Without setting ourselves the goal of detailing the theory of the above-mentioned phenomena, we will further restrict ourselves to only the most important points of this theory.

The transformer included in the circuit when the secondary circuit is open is in everything similar to a conventional reactive coil with iron. Suppose first that the reactive coil does not have an iron core, and the active resistance of its winding is negligible and can be neglected. In a steady state, the magnetic flux of the reactive coil changes according to the basic law of electromagnetic induction: v = w * dF / dt * 10-8, where v is the instantaneous value of the applied voltage, w is the number of coil turns, dF is the change during dt of the magnetic flow. The total change in the magnetic flux for any period of time t, counted from the zero value of the flux, will be expressed

Фt = 0 / tdФ = 0 / tvdt / w × 10-8

The flux Фt is a flux that permeates the coil at time t. It is an integral function of the applied voltage. Therefore, if the voltage changes along a sinusoidal curve, then the magnetic flux will also change along a sinusoidal curve with a phase shift of 1/4 period. In an unsteady turn-on mode, the magnetic flux of the reactive coil changes according to the same basic law of electromagnetic induction, but the shape of the curves of its change in time depends on the moment the coil is switched on to the primary network. Suppose that the primary voltage changes along the sinusoidal curve V1 and the coil is turned on at the moment the voltage passes through the largest value (Fig. 1a).

At the first moment of switching on, the magnetic flux is zero, but it will immediately begin to grow along the curve, which is the integral curve of the voltage V1, in Figure 1a along the curve Фy. Starting from zero value, the flux will increase as long as the voltage has a positive value, i.e. until the moment of zero voltage value, at which point the magnetic flux will reach its maximum value and will already begin to decrease. It is quite clear that the change in the magnetic flux will in this case occur along the same sinusoidal curve as in the steady state with a lag from the voltage by 1/4 of the period. Since it is assumed that there is no iron in the reactive coil, the magnetizing current in its change will coincide in phase with the magnetic flux, i.e. will change along a sinusoidal curve coinciding with the flow curve in fig. 1a along the curve iy. In view of the fact that the increase in the magnetic flux during the considered turn-on is the same as in the steady-state mode, the turn-on current will be equal to the steady-state current. Suppose now that the coil is switched on at the moment the voltage passes through zero (Fig. 1b). From this moment on, the magnetic flux of the coil will increase until the applied voltage becomes equal to zero, that is, during a half-period. The increase in flux will stop at the moment the voltage passes through zero. During the next half-period, the flux will decrease until the voltage changes its direction.

The change in the magnetic flux in this case is shown in Fig. 1b curve Ф, and the change in the magnetizing current curve i. Since the curve of the magnetic flux Φ during the period of its growth is an integral curve for a half-period, and not for 1/4 of a period, as in the first switch-on, it is clear that the largest value of the ordinate of the curve Φ, and hence of the curve i, is twice the corresponding values of the ordinates curves Фy and iy. This means that the turn-on current of a reactive coil without iron at the moment the voltage passes through zero is twice as much as the turn-on current of the same coil at the moment the voltage passes through the largest value. If the active resistance of the coil were really zero, then the magnetic flux and, therefore, the current pulsed would be indefinitely long without changing its signs, i.e. the current in the coil circuit would be pulsating in a constant direction. The magnetizing current i (as well as the magnetic flux Ф) in the considered case of switching on, we can imagine as if the sum of the current iy of the steady-state mode and the direct current ip equal to the largest value of the steady-state current Iy (Fig. 2), i.e. when the coil is turned on, a direct current ip is superimposed on the steady-state magnetizing current iy. Our assumption is that the active resistance (Fig. 2). coil is zero, not the same as reality. The presence of an active resistance quickly reduces the direct current Ip to zero, as a result of which the turn-on current gradually turns into a steady state current. In the case when the coil is turned on at an intermediate moment between the highest and zero voltage values, the turn-on current curve in its form occupies an average position between the current curves of the considered turn-on cases. Figure 3 shows the curve of the turn-on current i under the assumption that the turn-on occurred after a period of time t after the passage of the voltage V1 through zero, and that the resistance of the coil is not equal to zero. It is easy to see that the turn-on current in this case is no longer pulsating, constant in direction, but it is not an alternating symmetrical steady-state current either. We can consider this current as the result of the addition of two currents: the steady-state current, changing along the curve iy, and the constant current in the direction, decreasing along the curve in. The sums of the ordinates of the curves iy and in give the ordinates of the curve i. The values of the currents iy and ip, as well as the time during which the direct current disappears and the turn-on current turns into a steady current, depends on the value of the active resistance of the coil R and the self-induction coefficient L.

A real transformer connected to the primary network with no load differs from the considered reactive coil in that it has a very high self-induction coefficient and has an iron core. The presence of iron significantly increases the turn-on current. Indeed, let the switch-on occur at the moment the voltage passes through zero. In this case, the magnetic flux should increase to double its steady state value. Consequently, the induction in the iron should double, which will lead to its strong saturation and high magnetic resistance. The latter circumstance results in an excessive increase in the magnetizing turn-on current.

In modern transformers, especially with artificial cooling, the magnetic circuit is taken with a high saturation, and therefore the inrush currents when turned on must be large. Oscillograms of the turn-on currents of modern transformers show that the current surges exceed the amplitude of the normal magnetizing current by a factor of 100-120. Since the normal magnetizing current is 5-10% of the normal load current, the inrush current at turn-on can exceed the normal load current by 8-12 times. Such currents are dangerous for devices included in the transformer circuit, and are undesirable for the network to which the transformer is connected. They are also undesirable for the transformer itself due to the mechanical forces that are obtained between the winding coils. Due to their short duration, these currents are not thermally dangerous. To illustrate what has been said about turning on the transformer, Fig. 4 shows oscillograms of the turn-on currents of one transformer, and the first oscillogram corresponds to the case of turning on when the voltage passes through the largest value, i.e. through its amplitude, and the second oscillogram is the case of switching on when crossing zero. In order to weaken the switching current, switches with so-called preliminary contacts are used, with the help of which, at the first moment, a large resistance is introduced into the transformer circuit, which is short-circuited with further movement of the knife of the switch. In addition to the transient current phenomenon, transient voltage phenomena occur when the transformer is turned on, which often lead to an excessive voltage increase between adjacent turns of the winding and between the transformer terminals. The reason for these phenomena lies in free oscillations arising in a circuit consisting either of the line capacitance and self-induction of the transformer itself, when the latter is switched on with the secondary winding connected to the line, or from the capacitance of the transformer itself and its self-induction, when one higher voltage winding is switched on, which has quite large capacity. By mathematical analysis of free oscillations, it is easy to show that these oscillations can be considered as the resultant of traveling waves with a steep front, moving along the circuit in opposite directions at a very high speed, and their mutual shift and the height of the front depend on when the voltage changes the transformer is turned on. A traveling wave, moving along the winding of the transformer, gives a voltage between the turn above which the wave front is at a given moment and the next turn, which significantly exceeds the voltage that exists between the turns in the steady state. If the transformer is turned on at the moment the mains voltage passes through the largest value (amplitude), then the height of the wave front, and therefore the voltage between adjacent turns, can reach the value of the mains voltage amplitude, i.e., dozens of times exceed the normal voltage between turns, equal to V / w, where w is the number of winding turns.

For low voltage transformers, which have a large margin of dielectric strength of the insulation compared to the voltage being serviced, such an overvoltage between the turns is not dangerous. It is dangerous for high voltage transformers, in which the insulation works closer to the breakdown voltage. A means of combating breakdowns from local overvoltages is to strengthen the insulation of the first turns of the winding and turn on a reactive coil in front of the winding. In addition to local overvoltage, traveling waves, under favorable conditions for switching on, can also give an overvoltage at the winding terminals, reaching double the normal voltage.

The phenomena when the transformer is turned off largely depend on the conditions of the circuit break, namely: on the state and type of contacts of the switch, on the speed of the circuit breaking, on the environment in which the circuit is broken, etc. The time of the disappearance of the current in the circuit depends not only on the speed of divergence of the switch contacts, but also on the speed at which the volt arc formed between the diverging contacts extinguishes. In poorly designed circuit breakers, after the actual breaking of the circuit, the current in it is maintained for several more periods through the volt arc. The presence of a volt arc, causing free oscillations, can lead to significant overvoltage in high voltage transformers, the capacity of the windings of which is quite large. However, even in the case when the shutdown is not accompanied by a noticeable volt arc, for example, in oil switches, a large overvoltage can result in the winding, this time due to a rapid decrease in the current, that is, due to the significantly exceeding the normal value of the ratio di / dt; a rapidly decreasing field induces in this case a large voltage in the transformer winding.

Dangerous overvoltages arise in the transformer, and then when it, being connected to only one generator (and not to the station buses), will immediately be turned off at full load from the high voltage side, that is, from the secondary circuit. The fact is that modern transformers, even at their normal voltage, operate with rather strong saturation, if this voltage rises significantly, as is the case in this case when the load is dropped from the generator, especially the turbine generator with its almost rectilinear magnetization curve, then the saturation the magnetic circuit of the transformer will increase to a very large extent, and this will lead to a strong distortion of the magnetizing current curve, i.e. to the appearance of higher harmonics in it. These harmonics will cause oscillations of all kinds of frequencies in the generator and transformer circuits, up to the highest, which can significantly increase the voltage amplitude, i.e. cause overvoltage at the terminals of the transformer windings. In addition, due to the high frequency, local oscillations may occur in the winding, entailing damage to the insulation between the turns. Overvoltage at the transformer winding also appears when a long line or cable is turned off without load. It is a consequence of the secondary switching on of the line, which occurs through the reappearing voltage arc between the already diverged contacts of the switch. This re-ignition of the arc is explained as follows. When the line is turned off at the moment the current passes through zero - and at this very moment the oil switches are usually turned off - the voltage just passes through its amplitude (because the load of a long line is almost capacitive); this voltage will remain at the disconnected ends of the line as a charging voltage. The voltage of the transformer winding will continue to change in a sinusoidal manner. After half a period between the contacts of the switch on the side of the transformer and the contacts of the switch on the side of the line, a double normal voltage will act, which can cause a volt arc and, as it were, a secondary turn on of the line, but already with a double voltage. This connection of the line will give waves traveling in opposite directions with a front of double, in comparison with normal switching, heights, and, consequently, dangerous overvoltages for both the line and the transformer. With poor contact arrangement or slow shutdown, the arc can be re-ignited several times. The imperfection of the circuit breaker, namely, the non-simultaneous switching on of all phases, damage to one of the contacts or the breakage of one or two phases of the line - can also give an overvoltage at the transformer located at the end of the line. Indeed, if one line is open, then from the self-induction of the transformer, the capacitance of this open line and the capacitance of the rest of the line connected in series with it (Fig. 5), a circuit is formed in which free oscillations of the same frequency as the frequency of the supply current. As a result of this, a voltage resonance will appear in the circuit, and, consequently, an overvoltage of the transformer winding, reaching a large value.

Dangerous overvoltages arise in the transformer, and then when it, being connected to only one generator (and not to the station buses), will immediately be turned off at full load from the high voltage side, that is, from the secondary circuit. The fact is that modern transformers, even at their normal voltage, operate with rather strong saturation, if this voltage rises significantly, as is the case in this case when the load is dropped from the generator, especially the turbine generator with its almost rectilinear magnetization curve, then the saturation the magnetic circuit of the transformer will increase to a very large extent, and this will lead to a strong distortion of the magnetizing current curve, i.e. to the appearance of higher harmonics in it. These harmonics will cause oscillations of all kinds of frequencies in the generator and transformer circuits, up to the highest, which can significantly increase the voltage amplitude, i.e. cause overvoltage at the terminals of the transformer windings.

Thank you very clearly explained the current surges when the transformer was turned on.

Why is the transformer windings discharged after each experiment?

After turning off, the trance discharges itself very quickly. You confused the trance with a capacitor that stores a charge for months.

If all the wires are disconnected from it, can there still be current in it?

No