PROCEEDINGS
OF THE TOMSK ORDER OF THE OCTOBER REVOLUTION
AND ORDER OF THE RED BANNER OF LABOR
POLYTECHNIC INSTITUTE named after S. M. KIROV

Volume 262 1973

TWO-PHASE RELAY SYSTEM FOR REGULATING THE STARTING CURRENT OF TRACTION MOTORS

A. P. ZAITSEV, V. A. PODLYAGIN

(Presented to the scientific and technical seminar of the Research Institute of Automation and Electromechanics)

Relay control systems, in comparison with pulse-width systems, have primarily two advantages: high speed of response and high accuracy in maintaining the mean value of the controlled quantity, independent of the nature and magnitude of disturbing factors. However, relay systems constructed according to the classical principle are used very rarely for the purpose of regulating the starting current of traction electric motors because of a number of disadvantages inherent in them. The first disadvantage is due to technical reasons and consists in significant pulsations of the controlled current when controllers are made on the basis of standard elements possessing inertia, delay, and ambiguity of static characteristics. The second disadvantage is due to the fundamental properties of relay systems and manifests itself in a variable control frequency, which is determined by the nonlinear parameters of the system and the nature of the disturbing factors. The third disadvantage follows from the second and consists in the fact that in ordinary relay automatic control systems it is impossible to realize the principle of multiphase control.

The first disadvantage is eliminated by developing special elements: fast-response current sensors, hysteresis-free high-sensitivity comparison devices, and functional elements with insignificant delay. The last two disadvantages are of no substantial importance when traction electric motors are supplied from autonomous sources, and are the main obstacle to applying the relay principle of control in electrically propelled installations supplied from a contact network. In the latter case, the variable control frequency leads to two undesirable consequences: to an effect on track automation devices tuned to a definite frequency, and to difficulties in filtering the low frequencies that occur at the end of the control process.

An ideal relay automatic current-control system for traction motors would have a constant switching frequency. Such an automatic control system would combine the positive properties of the relay and pulse-width principles of control, i.e., it would possess high speed of response and accuracy of control, a constant switching frequency, and would allow the application of the principle of multiphase control. Multiphase control ensures a high frequency of current pulsations in the network and makes it possible substantially to reduce the dimensions of the input filter.

Fig. 1. Block diagram of the two-channel relay control system.

Visible labels in the diagram: \(+\), \(-\), \(TK_1\), \(TK_2\), \(D_1\), \(D_2\), field winding \(OV_1\), field winding \(OV_2\), \(I_1\), \(I_2\), \(U_1\), \(U_2\), \(U_3\), \(U_4\), \(U_5\), \(U_6\), \(U_7\), \(U_k\), \(U_{10}\), \(FEM\), \(FBM\), \(BF\), \(SZ\), \(SS\), \(DPT\), \(MV\), Channel I, Channel II.

Below, the principle is considered for constructing a relay automatic control system for the current of a traction motor, ensuring a constant switching frequency under two-phase regulation. A feature of such an automatic control system is the forced turn-off, at a constant frequency, of the regulating thyristor switch.

The turn-off frequency of the thyristor switch determines the frequency of motions in the system and must be chosen in a definite way: it

Fig. 2

must be no higher than the initial possible regulation frequency for the specified current setting. If the turn-off frequency is chosen above the indicated value, the parameters of the automatic control system do not ensure initial motion in the system at such a frequency, and the principle of operation will be violated.

The block diagram of the system is given in Fig. 1, and the diagrams explaining its operation are shown in Fig. 2. The armature current \(I_1\) is converted by the armature-current sensor DTA into the voltage \(U_d\), which is compared by the comparison circuit CC with the setting voltage \(U_0\). If \(I_1\) is less than the setting current, then CC generates voltage pulses \(U_k\), which, through the de-

of comparison circuit CS are amplified in power and increased in duration by the blocking former BF.

The output voltage pulses \(U'_4\) of the blocking former are amplified by the power amplifier \(Y_1\) and open the thyristor switch TK\(_1\). The armature current \(I_1\) increases. Closing of TK\(_1\) is carried out forcibly and periodically as follows. The magnetic-coupled multivibrator MB, by the leading edge of the output voltage \(U_1\), starts the blocking-pulse former BPF, which at one of its outputs, by a rectangular voltage pulse \(U_2\), acts on the CS, prohibiting the passage of signals from the CC for the duration of the pulse, and at the second input starts the delayed-pulse former DPF. The latter forms a short voltage pulse \(U'_3\), located in time at the center of the pulse \(U_2\). After amplification by amplifier \(Y_2\), the voltage pulses \(U'_3\) are fed to the quenching thyristor TK\(_1\), causing a decrease in the armature current.

The blocking pulses \(U_2\) exclude the simultaneous appearance, or inadmissibly close placement, of unlocking and locking pulses, thereby completely excluding failures of TK\(_1\). The duration of the blocking pulses is chosen equal to twice the recharge time of the commutating capacitor of TK\(_1\).

Thus, in the system only the turn-on instants of TK\(_1\) are monitored by the comparison circuit, and in steady-state modes the system operates as a pulse-width system, although it has no pulse-width modulator. In transient processes the relative on-duration \(\gamma\) can change stepwise from zero to some value \(\gamma_{\max}\), which is determined by the duration of the blocking pulses \(U_2\) and can be defined as

\[ \gamma_{\max}=1-\frac{t_{\mathrm{B}}}{2} f \cdot 10^{-6}, \]

where

\(t_{\mathrm{B}}\) (\(\mu\)sec) is the duration of the blocking pulses;

\(f\) (Hz) is the switching frequency.

The ability of the system to change \(\gamma\) stepwise within the indicated limits provides it with high speed of response.

The second control channel operates in an analogous manner, but all processes in it are shifted by 180 electrical degrees.

Figure 2 also shows the waveforms of the current \(I_{\phi}\) consumed by the motors from the capacitor of the input filter, and the current \(I_c\) consumed by the system from the network.

The use of such a system makes it possible to double the ripple frequency of the line current and substantially reduce the dimensions of the input filter.