Pulse Speed Control Systems for Wound-Rotor Induction Motors

A. I. Zaitsev, Yu. P. Kostyukov

(Recommended by the scientific seminar of the Electromechanical Faculty)

The principal direction of technical progress in the field of automated electric drives is becoming the use of semiconductor technology and, above all, thyristor devices—the most economical and possessing increased reliability \[1\].

The development of semiconductor technology makes it possible to reconsider the question of using the pulse method for controlling the speed of wound-rotor induction motors. When this control method is applied, the speed can be regulated over fairly wide limits and the required mechanical characteristics can be obtained. Its great advantage is the simplicity of the control circuits and their fast response. Pulse systems quite fully satisfy the requirements of certain general-purpose industrial mechanisms (pumps, fans, compressors, etc.) with respect to the depth and smoothness of speed control.

As is known, speed control of wound-rotor induction motors is carried out by changing the resistance in the rotor circuit and by cascade circuits, in which the slip energy is returned to the network through a converter device of the valve or machine type \[2\]. The pulse method of speed control can be applied in both cases. In this arrangement, an artificial switching circuit using thyristors is introduced into the direct-current rotor circuit in parallel with the converter device or the additional resistance. For purely valve systems, such a circuit has certain advantages from the standpoint of the power factor of the supply network \[3\]. The energy recuperated to the network depends on the control range and the nature of the load. The converter device may be common to several cascades.

For circuits with pulse-controlled resistance, the principal disadvantage of the rheostat method—the stepwise nature of regulation—is eliminated.

With the pulse method, speed control is carried out by changing the relative duration of switching-on of the power thyristor, which is equal to

\[ \gamma=\frac{T-t_u}{T}, \]

where: \(T\) is the pulse repetition period;
\(t_u\) is the pulse duration, corresponding to the time during which the converter (or the additional resistance) conducts current.

In pulse speed control of wound-rotor induction motors, it is most advisable to use a commutation law with a constant off-state time of the power thyristor and a variable frequency [4].

To reduce current pulsations, it is desirable to choose \(t_u\) as small as possible, guided by the limiting value of the operating frequency of the power thyristor. Since commutation losses in thyristors are directly proportional to frequency and current, in order to obtain small values of \(\gamma\) under a fan-type load it is quite permissible to overstate the operating frequency of the power thyristor in comparison with its rated value. Usually \(t_u\) lies within the range \(500 \div 2000\ \mu\text{s}\), with the larger values applying to drives with constant load.

Figure 1 shows the simplest circuit, in which a pulse-controlled resistance is used to control the speed of a wound-rotor motor (IM). Choke \(P\) serves to smooth current pulsations in the rotor circuit.

Fig. 1.

The lower limit of speed control at a given load is determined by the value of the additional resistance \(R_d\). At \(\gamma=0\), the rotor current is closed through \(R_d\), and the slip energy is dissipated in this resistance.

Operation at high speed is achieved by increasing the relative on-time. At \(\gamma=1\), the rotor of the induction motor, through the rectifying three-phase bridge \((B)\), is short-circuited by the power thyristor of the artificial-commutation circuit \((IB)\). The residual slip of the motor is determined by the voltage drop in its rotor, in the rectifier bridge, and in the thyristor. The mechanical characteristic nearly coincides with the natural one; the installation operates at the highest efficiency.

For arbitrary relative durations \((0<\gamma<1)\), the artificial-commutation circuit makes it possible, as it were, to vary smoothly the value of the active resistance from \(R_d\) to 0.

Speed control by introducing resistance into the rotor circuit is uneconomical and can be used only for motors with a power of up to 100 kW.

Artificial-commutation circuits using thyristo-

tors. The energy indicators and operating reliability of pulse systems depend on the correct choice of the commutation frequency and of the artificial turn-off circuit.

A wound-rotor induction motor can operate in steady state under short-term overloads and during transients, consuming a considerable starting current; operation of the motor in an intermittent-current mode is also possible; to extend the upper and lower ranges of regulation, limits of slip variation on the order of 0.05–0.95 are required. At large currents and low voltages, the value of the commutating capacitor proves to be considerable. All this imposes certain requirements on artificial turn-off circuits. Obviously, preference should be given to circuits in which the energy stored in the commutating capacitor increases with increasing load current, which greatly facilitates the process of turning off the power thyristor. This is achieved by inserting an inductance \(L_1\) operating in the intermittent-current mode. In the presence of inductance \(L_1\), the capacitor is charged to a voltage greater than the average value of the rectified voltage \((\gamma = 0)\). The value of this inductance is chosen to be small (tens of microhenries) in order to limit the maximum forward voltage on the power thyristor.

The capacitance of the capacitor is determined by the maximum value of the current to be commutated, the voltage on the capacitor, and the recovery time of controllability of the power thyristor \((T_1)\).

Below, several circuits are considered, differing in the method of recharging the capacitor and in its connection to the terminals of the power thyristor to ensure that the latter is blocked. To simplify the drawings, the control circuits and the motor are omitted.

Fig. 2.

Figure 2 shows an autotransformer circuit of artificial turn-off [5, 6].

We shall assume that thyristor \(T_2\) has been turned on. When \(T_1\) is turned on, the lower plate of the capacitor charges resonantly through \(L_2\) to the maximum positive voltage, which is held by the blocking diode \(D_1\). Meanwhile, the rotor current passes through \(T_1\) and \(L_1\). After \(T_2\) is turned on, thyristor \(T_1\) is biased in the reverse direction and is blocked; the capacitor is recharged and simultaneously takes up the load current; the artificial turn-off circuit returns to its initial state, and so on.

The autotransformer circuit is analogous in its operating principle to the circuit described in [7], except that in the latter there is no magnetic coupling between the inductance coils.

Since the capacitor is charged above the average value of the motor voltage at \(\gamma = 0\), the capacitor may discharge through \(D_1\), \(L_2\), \(L_1\), and the converter or through an additional resistance, which are not shown in the figure. The reduction of the initial capacitor voltage depends on the parameters of the discharge circuit and may be insignificant; otherwise, the effectiveness of the commutation circuit drops sharply. This drawback is most easily eliminated by replacing the blocking diode \(D_1\) with a thyristor \(T_3\) operating synchronously with \(T_1\).

Figure 3 shows a modernized autotransformer circuit in which the size of the power thyristor does not have to be overestimated, since recharging of the capacitor occurs only through \(T_3\). Short-circuiting of the load current simultaneously through \(T_3\) and \(T_2\) is practically eliminated.

Obviously, these circuits are expedient to use at frequencies up to 500 cps; in this case the elements of capacitor commutation can have small dimensions.

Fig. 3.

Fig. 4.

The main drawback of the circuits considered, and of analogous ones [5], is that it is difficult to obtain a small \(\gamma\) at a high commutation frequency, since the capacitor \(C\) is recharged while the power thyristor \(T_1\) is conducting.

The circuit of Fig. 4 [8] is to some extent free of these shortcomings; in it the capacitor \(C\) is charged through the valve \(D_1\). After \(T_2\) is switched on, an oscillatory process arises in the \(L_2\)–\(C\) circuit. When recharging of the capacitor is completed, the valve \(D_1\) begins to conduct current and both thyristors turn off. The half-period of the natural oscillations of the circuit should not be greater than 30–40 μsec; otherwise the size of the capacitor has to be increased [8].

The rated power of the auxiliary thyristor is chosen on the basis that the greatest peak of the current passing through \(T_2\) exceeds twice the load-current value. Although \(T_2\) conducts current for a short interval of time, its size and cost will be large at a modulator frequency of 1000 cps and a constant load.

Other modifications of the circuit of Fig. 4 are also possible. Thus, by a combination of auxiliary thyristors and saturable chokes, more economical circuits are obtained at frequencies of 1 kcps and less [8].

The proposed circuit (Fig. 5) differs advantageously from the preceding ones in the smaller dimensions of the capacitor-commutation elements at a frequency of 1 kcps and higher. When a triggering pulse is applied to \(T_2\), the capacitor \(C\) is charged to a voltage of the indicated polarity. Resonant recharging of the commutating capacitor occurs through the inductance coil \(L_2\) and the blocking

Fig. 5.

diode \(D_1\) during the interval of the off state of the power thyristor \(T_1\). With the arrival of the next pulse at \(T_2\), capacitor \(C\) discharges and, for approximately the duration of the recovery time of the controllability of the power thyristor, takes upon itself the motor current; after this the capacitor is charged and all processes are repeated. In this circuit, definite requirements are imposed on the value of the inductance \(L_2\), which is selected from the condition that the half-period of the natural oscillations of the recharging circuit not exceed the minimum duration of the pulse-repetition period and be sufficient for recovery of the controllability of the auxiliary thyristor. Thus, this circuit is applicable only under the law of commutation with a constant duration of the off state of the power thyristor and with variable frequency. In the circuit of Fig. 5 it is easy to obtain the required minimum \(\gamma\).

The considered artificial turn-off circuits (Figs. 2–5) provide reliable commutation in the region of continuous rotor current and are not intended for operation in the region of discontinuous currents. The circuit with charging of the capacitor from an additional source is free of this drawback (Fig. 6). Capacitor \(C\) is charged through inductance coil \(L_2\) and blocking diode \(D_1\) to approximately twice the voltage of the supply source. The minimum duration of the pulse-repetition period must, with a small margin, exceed the half-period of the natural oscillations of the charging circuit. When \(T_2\) is fired, capacitor \(C\) discharges through inductance \(L_3\), and a positive voltage appears at the cathode of \(T_1\), which leads to the blocking of the power thyristor. The auxiliary thyristor is blocked when the current in the oscillatory circuit \(T_2\)—\(L_3\)—\(C\) changes its direction. The condition for normal operation of the circuit is that the amplitude of the current in the oscillatory circuit must exceed the maximum value of the current through the power thyristor. To obtain a short recovery time for the power thyristor, it is necessary that a sufficiently high reverse current flow through the device during the recovery period. For this purpose, a capacitor \(C_1\) of sufficiently large capacitance (\(C_1 \ge C\)) is connected in parallel with the converter device (not shown in Fig. 6). The reverse-current pulse passes in the circuit \(T_2\)—\(T_1\)—\(D\)—\(C_1\) for several microseconds. In this case the elements of the commutating circuit have small dimensions in comparison with the circuit without \(C_1\). The main drawback of the circuit of Fig. 6 is the presence of an additional source.

Fig. 6.

In conclusion, we note that the artificial turn-off circuits considered can also be used for other purposes of pulse control.

LITERATURE

1. Fourth All-Union Conference on Automated Electric Drive. Resolution. VNIIEM, 1965.

2. D. A. Zavalishin, V. A. Prozorov. Basic trends and prospects for the development of adjustable AC electric drives. In the collection: Automated Electric Drive, Nauka, 1964.

1. K. Neumann. Pulse Control of D-C and A. C. Motors by Silicon-Controlled Rectifiers. IEEE Transactions on Communication and Electronics, 1964, V. 83, No. 73.

2. A. I. Zaitsev, Yu. P. Kostyukov. An asynchronous converter cascade with pulse control. Izvestiya TPI, vol. 153, 1965.

3. Silicon-controlled rectifier thyristors, technical handbook (translated from English), Energiya, 1964.

4. Neville W. Marham, John C. Hey. The control of Battery Powered DC Motors Using SCR’S in the Jones circuit. IEEE International Convention Record, 1964, No. T-4.

5. Gurwicz D. A pulsed d. c. motor control system. Electrical Review, 1964, No. 175, No. 19.

6. R. E. Morgan. Time Ratio Control with combined SCR and SR Commutation. IEEE Transactions on Communication and Electronics, 1964, No. 83, No. 73.