DEVELOPMENT AND STUDY OF THE MUS-D ELECTRIC DRIVE WITH A CLASS D PPU

V. A. SEVASTYANOV, A. P. INESHIN, A. P. RYBAKOVA
(Recommended by the scientific seminar of the Department of Electric Drive of the Tomsk Polytechnic Institute)

In recent years, domestic and foreign machine-tool plants and design organizations have been moving toward the use of packaged drives, with considerable attention being paid to the use of self-saturating magnetic amplifiers (MUS). Among domestic drives of this type, in the speed-control range of the actuator motor of 10:1 and 100:1, the most widely known and broadly used in industry are the PMUM drives developed by ENIMS [2]. As shown in [6, 8] and in other studies, the development of similar drives of a wider range, with the required stiffness of mechanical characteristics, dynamic performance, and operating reliability, necessitates the creation of fundamentally new circuit solutions for regulators for MUS-D systems and methods of stabilization.

On the basis of studies carried out on a number of existing drives [6, 2] and the recommendation of ENIMS, the Department of “Electric Drive and Automation of Industrial Installations” of the Ulyanovsk Polytechnic Institute developed and studied, in the institute’s industrial laboratory, a reversible direct-current electric drive using the MUS-D system with a class D PPU, with a speed-control range of 200:1, meeting the above requirements, for the carriage-feed mechanism of a carousel-type lathe.

The electrical circuit of this drive is shown in Fig. 1. Here ID is the actuator motor; SPT is a static current converter for supplying the motor, consisting of a three-phase rectifier and an MUS with an alternating-current output, connected according to a circuit with external loop feedback; PPU is an intermediate semiconductor class D amplifier, used as a preliminary amplification stage and including a pulse-width modulator (PWM), a pulse amplifier (PU), and an output stage (K); ZS is a speed setter with a variable transmission coefficient \((m)\), implemented on the basis of the standard PP-36 switch of the NVA plant, in the form of a high-resistance discrete potentiometer (3 kΩ), and ensuring constancy of the static error over the range as a function of the speed setting by introducing additional into the input circuit of the PDU

Fig. 1.

resistances \(R_a\) [7]; TG is a TMG-30P tachogenerator, used as a speed sensor. As the static current converter, with minor modification, a standard PMU-9M-1 ENIMS power unit, mass-produced by the Prokop’evsk “Elektromashina” plant, may be used.

The principal technical data of the specified electric drive are summarized in Table No. 1. Here the P-41 motor, in order to improve the dynamics and reduce the effect of tooth ripple, has been selected with a somewhat oversized power rating (6 kW instead of 4.5 kW).

The initial state of the circuit is determined by the zero position of the regulator; in this case \(U_z = 0\), and \(PT_1\) is open owing to the positive bias on its base through the circuit \(R_6\) and is in saturation. Therefore, when capacitor \(C_4\) (\(C_5\)) discharges, because of the low collector–emitter junction resistance, the voltage at its output is zero. The triode \(PT_2\), owing to the negative bias through \(R_{10}\), is open, and current flows in its emitter–collector circuit. The load of \(PT_2\) is \(R_{11}\), which at the same time supplies a negative bias to the base of \(PT_3\). However, because \(PT_2\) is open, the base of \(PT_3\) is at zero potential, which causes it to close. The triode \(PT_4\) is open because of the negative bias on its base through the circuit \(R_{12}\) and \(R_{13}\); therefore a bias current flows through the control winding (CW), which results in the minimum output of the MUSA (\(I_y = 0.3\) A), and the motor is at rest. When the regulator is moved to the operating position, a voltage \(U_z \ne 0\) of polarity appears at the input of the SPU: negative at the base and positive at the emitter of \(PT_1\). This causes \(PT_1\) to close, diode \(D_1\) to break down, and positive periodic voltage pulses with a frequency of 100 Hz to be applied to the base of \(PT_2\). A series of negative pulses at the output of \(PT_2\) is inverted and amplified in \(PT_3\), which leads to the appearance, at the base of \(PT_4\), of pulses of positive polarity of the same frequency, causing a decrease in the bias current and the opening of the MUSA. Thus, the rectangular output voltage of the IU and its relative duration \(\gamma\) control the operation of the output stage of the amplifier—the semiconductor switch \(K\)—in the load circuit (CW). The average value of the current in the control winding, connected into the supply circuit for the time \(T^+ = \gamma T_k = K_1 U_y\) during the switching period \(T_k\), is proportional to the input signal \(U_y\). Rotation of the electric motor gives rise to \(E_{\mathrm{tg}}\) and an MUSA output proportional to \(U_y = (U_z - E_{\mathrm{tg}})m\), in accordance with the speed setting. An increase in the load on the motor shaft causes a decrease in \(n\) and \(E_{\mathrm{tg}}\), which leads to an increase in \(U_y\), a more negative potential at the base of \(PT_1\), an increase in the MUSA output, and restoration of \(n\) to the prescribed value.

To reduce the time constant of the CW and increase response speed, an additional resistance \(R_{16}\) is connected in series with it. Protection of \(PT_4\) against overvoltages on the control winding that arise during transients is provided by diode \(D_{11}\). Breakdown of \(PT_4\) (across the emitter–collector junction) or an open circuit in the SPU input circuit results in zero output of the MUSA at any speed, since in this case the maximum bias current flows through the CW. Diodes \(D_5\) and \(D_6\) protect the SPU input against overvoltages.

Stabilization of the electric drive is carried out electromagnetically, by the derivative of the total magnetic flux of the MUSA. To pick off the correction signal, the free control winding OK-MUS [1] is used. The correction signal is characterized by a low content of interference, the effect of which is weakened by introducing a simple integrating circuit.

Table No. 1

The circuit provides a current-limiting unit, implemented on the basis of current transformers and a reference Zener diode $D_{12}$. When the armature current exceeds the permissible value, $D_{12}$ breaks down and a signal is applied to the PPU input (to the resistance $R_5 = 150$ ohms), limiting the current at the MUS output. Owing to the known characteristics of a static current converter based on an MUS, electric braking is absent in the circuit. Reversal and rapid braking are carried out by means of electromagnetic clutches, widely used in machine tools.

Data on the structural design, the product nomenclature, a description of the operation of the indicated drive units, and theoretical studies are given in [7] and in other articles by the authors.

Fig. 2.

The block diagram of the developed drive system is shown in Fig. 2. Considering the operation of the automatic-control system “in the small” [5], the following values of the transfer functions may be adopted:

\[ W_5(p)=\frac{\dfrac{1}{R_{\mathrm{e}}}}{T_{\mathrm{e}}p+1}; \qquad W_6(p)=\frac{\dfrac{R_{\mathrm{e}}}{C_{\mathrm{e}}}}{T_{\mathrm{m}}p}; \qquad C_{\mathrm{e}}=\frac{U_{\mathrm{n}}-I_{\mathrm{yan}}R_{\mathrm{ya}}}{n_{\mathrm{n}}}. \]

The transfer function of the motor is:

\[ W_{\mathrm{e}}(p)= \frac{\dfrac{1}{C_{\mathrm{e}}}} {T_{\mathrm{e}}T_{\mathrm{m}}p^2+T_{\mathrm{m}}p+1}, \tag{1} \]

where $T_{\mathrm{m}}$, $T_{\mathrm{e}}$ are, respectively, the electromechanical and electromagnetic time constants of the electric motor, taking into account the internal resistance of the MUS referred to the motor.

\[ T_M=\frac{GD^2 R_{\mathrm{e}}}{375\,C_e C_M}; \qquad T_{\mathrm{e}}=\frac{L_{\mathrm{a}}}{R_{\mathrm{a}}}; \qquad R_{\mathrm{e}}=R_{\mathrm{f}}+R_{\mathrm{a}} \]

\(R_{\mathrm{f}}\) — fictitious internal resistance of the MUS;
\(L_{\mathrm{a}}\) — inductance of the armature circuit.

The transfer function of the MUS, without allowance for the delay, the influence of which is considered in another paper by the authors:

\[ W_{\mathrm{MU}}(p)=\frac{K_{\mathrm{MU}1}\cdot K_{\mathrm{MU}2}}{T_{\mathrm{MU}}\cdot p+1}, \tag{2} \]

where \(T_{\mathrm{MU}}\) is the time constant of the MUS;

\[ K_{\mathrm{MU}1}=\frac{1}{R_{\mathrm{ov}}+R_{16}};\qquad K_{\mathrm{MU}2}=K_1R_{\mathrm{e}} \]

are the gain factors of the first and second stages of the MUS;

\(K_1\) — the current gain factor of the MUS.

The introduction of stabilizing feedback, as adopted in the circuit of Fig. 1, is equivalent to the use of a differentiating transformer [1]; therefore the transfer function of the correction element may be represented in the form:

\[ W_{\mathrm{k}}(p)=\frac{e_{\mathrm{k}}(p)}{I_{\mathrm{ov}}(p)} =\frac{K_{\mathrm{k}}p}{T_{\mathrm{k}}p+1}, \tag{3} \]

where

\[ K_{\mathrm{k}}=\frac{M\,(R_{\mathrm{in}}+R_{12})}{R_{\mathrm{ok}}+R_{\mathrm{in}}+R_{12}} \]

is the transfer coefficient of the correcting element \([\Omega\cdot\mathrm{s}]\);

\(T_{\mathrm{k}}\) — the time constant of the correcting circuit, dependent on the parameters of the MUS;

\(M\) — a coefficient taking into account the presence of mutual inductance between the windings of the MUS.

According to Fig. 2, the voltage gain factor of the PPU, the transfer coefficient of the tachogenerator, and the input circuit of the PPU [7] are written as:

\[ W_2(p)=K_{\mathrm{p}}=z;\qquad W_9(p)=\gamma;\qquad W_1(p)=m=\frac{R_{\mathrm{in}}}{R_{\mathrm{in}}+R_{\mathrm{tg}}+R_{\mathrm{d1}}+R_{\mathrm{dk}}}. \]

Using the methods of transformation of block diagrams known in the linear theory of automatic control, we obtain the transfer function of the closed-loop ACS with respect to the control action. Substituting into it the values of the transfer functions of the ACS elements and, in view of the smallness, neglecting \(T_{\mathrm{k}}=0\), after simple transformations we obtain:

\[ W(p)= \frac{m\cdot K_{\mathrm{PPU}}\cdot K_{\mathrm{MU}1}\cdot K_{\mathrm{MU}2}\cdot K_{\mathrm{d}1}\cdot K_{\mathrm{d}2}} {a_0p^4+a_1p^3+a_2p^2+a_3p+a_4}. \tag{4} \]

Here

\[ a_0=T_{\mathrm{e}}\cdot T_{\mathrm{MU}}\cdot T_M\cdot T_{\mathrm{k}}=0, \]

\[ a_1=T_{\mathrm{e}}\cdot T_{\mathrm{MU}}\cdot T_M +K_{\mathrm{PPU}}\cdot K_{\mathrm{MU}1}\cdot K_{\mathrm{k}}\cdot T_{\mathrm{e}}\cdot T_M; \]

\[ a_2=T_{\mathrm{MU}}\cdot T_M+T_{\mathrm{e}}\cdot T_M +K_{\mathrm{PPU}}\cdot K_{\mathrm{MU}1}\cdot K_{\mathrm{k}}\cdot T_M; \]

\[ a_3 = T_{\mathrm{MU}} + T_{\mathrm{M}} + K_{\mathrm{PPU}}\cdot K_{\mathrm{MU}1}\cdot K_{\mathrm{k}}; \]

\[ a_4 = 1 + m\cdot K_{\mathrm{PPU}}\cdot K_{\mathrm{MU}1}\cdot K_{\mathrm{MU}2}\cdot K_{\mathrm{d}1}\cdot K_{\mathrm{d}2}, \]

\[ K_{\mathrm{d}1}=\frac{1}{R_{\mathrm{e}}}; \qquad K_{\mathrm{d}2}=\frac{R_{\mathrm{e}}}{C_{\mathrm{e}}}. \]

The drive stability condition according to Hurwitz, for \(a_1>0;\ a_2>0;\ a_3>0;\ a_4>0\), is written in the form:

\[ \left. \begin{aligned} &(T_{\mathrm{MU}}\cdot T_{\mathrm{M}} + T_{\mathrm{e}}\cdot T_{\mathrm{M}} + K_{\mathrm{PPU}}\cdot K_{\mathrm{MU}1}\cdot K_{\mathrm{k}}\cdot T_{\mathrm{M}})\times\\ &\quad \times (T_{\mathrm{MU}} + T_{\mathrm{M}} + K_{\mathrm{PPU}}\cdot K_{\mathrm{MU}1}\cdot K_{\mathrm{k}})>\\ &> (T_{\mathrm{e}}\cdot T_{\mathrm{MU}}\cdot T_{\mathrm{M}} + K_{\mathrm{PPU}}\cdot K_{\mathrm{MU}1}\cdot K_{\mathrm{k}}\cdot T_{\mathrm{e}}\cdot T_{\mathrm{M}})\times\\ &\quad \times (1+m\cdot K_{\mathrm{PPU}}\cdot K_{\mathrm{MU}1}\cdot K_{\mathrm{MU}2}\cdot K_{\mathrm{d}1}\cdot K_{\mathrm{d}2}). \end{aligned} \right\} \tag{6} \]

It follows from this that, for the specified drive parameters, the stability of its operation improves with an increase in \(K_{\mathrm{k}}\), which in practice can be achieved by increasing the number of turns \(W_{\mathrm{k}}\) of the correcting winding and by selecting the resistances of the stabilizing circuit.

An analytical expression for the mechanical characteristic without taking current limitation into account, in a linear approximation, can be obtained according to the method set forth in [4]. The equation of static equilibrium in the presence of a load on the shaft, as applied to the automatic control system under consideration in the closed state, can be represented as:

\[ \left[U_{\mathrm{z}}-\gamma\left(U_0-\Delta U_0-I_{\mathrm{a}}R_{\mathrm{e}}\right)C_{\mathrm{e}}^{-1}\right]\cdot m\cdot K_{\mathrm{PPU}}\cdot K_{\mathrm{MU}} =U_0+\Delta U_0, \tag{7} \]

where \(U_0\) is understood as the voltage taken from the fictitious “no-load characteristic” of the MUS [5], and \(K_{\mathrm{MU}}\) is the MUS gain factor.

Passing to speed, we can write:

\[ n=C_{\mathrm{e}}^{-1}\left(U_0+\Delta U_0-I_{\mathrm{a}}R_{\mathrm{e}}\right), \tag{8} \]

Solving (7) and (8) jointly, we obtain:

\[ n=C_{\mathrm{e}}^{-1}\left( \frac{U_{\mathrm{z}}\cdot m\cdot K_{\mathrm{PPU}}\cdot K_{\mathrm{MU}}} {1+K_{\mathrm{PPU}}\cdot K_{\mathrm{MU}}\cdot \gamma\cdot m\cdot C_{\mathrm{e}}^{-1}} - \frac{I_{\mathrm{a}}\cdot R_{\mathrm{e}}} {1+K_{\mathrm{PPU}}\cdot K_{\mathrm{MU}}\cdot \gamma\cdot m\cdot C_{\mathrm{e}}^{-1}} \right). \tag{9} \]

Assuming that the static coefficients entering into (9) vary only slightly over the range for the automatic control system under consideration, we have:

\[ \gamma=0.066;\quad K_{\mathrm{PPU}}=78;\quad K_{\mathrm{MU}}=56;\quad C_{\mathrm{e}}=0.09. \]

Therefore, for \(0.1<m<0.5\), which is the case in this circuit,

\[ K_{\mathrm{PPU}}\cdot K_{\mathrm{MU}}\cdot \gamma\cdot m\cdot C_{\mathrm{e}}^{-1}\simeq (300 \div 1500)\gg 1. \]

Then, neglecting unity in the denominator of expression (9), we obtain:

\[ n=\frac{1}{\gamma}\left(U_{\mathrm{z}}-\frac{I_{\mathrm{a}}R_{\mathrm{e}}} {K_{\mathrm{PPU}}\cdot K_{\mathrm{MU}}\cdot m}\right)=n_0-\Delta n. \tag{10} \]

It follows from (10) that, for \(I_{\mathrm{a}}=\mathrm{Const}\), \(\Delta n\) depends on \(m\). Let us determine how \(m\) should vary over the range in order to ensure constancy of static accuracy \((\Delta\%=\mathrm{Const})\) as a function of the speed setting, which is necessary when operating the feed drive of heavy machine tools [3]. For this purpose, in a linear approximation, one may use the expression:

\[ k=\frac{\Delta n_{\mathrm{r}}^{\%}}{\Delta\%}-1, \tag{11} \]

where: \(K\) is the static gain of the open-loop ACS required to ensure the necessary static accuracy \((\Delta\%)\);

\(\Delta n_p\%\) is the relative speed deviation in the open-loop ACS.

On the other hand, for the ACS under consideration in the open-loop state we have:

\[ K = m \cdot K_{\text{ppu}} \cdot K_{\text{mu}} \cdot \gamma \cdot C_e^{-1}. \tag{12} \]

Equating the right-hand sides of expressions (11) and (12), we obtain:

\[ m \cdot K_{\text{ppu}} \cdot K_{\text{mu}} \cdot \gamma \cdot C_e^{-1} = \frac{\Delta n_p\%}{\Delta\%} - 1. \]

Solving the last equality with respect to “\(m\)” as applied to operation of the drive at the lower limit of the speed-control range, when the shaft load varies from \(0.3 M_n\) to \(M_n\), and neglecting unity, we find the maximum value of the transmission coefficient of the input circuit of the PPU:

\[ m_{\max} = \frac{1}{K_{\max}} \cdot \frac{\Delta n_{pm}\%}{\Delta\%}. \tag{13} \]

Here,

\[ \Delta n_{pm}\% = \left( \frac{n_{0m} - 0.3 I_{yn} \cdot R_{em} \cdot C_e^{-1}}{n_{nm}} - 1 \right) \cdot 100\% \]

\[ K_{\max} = K'_{\text{ppu}} \cdot K'_{\text{mu}} \cdot \gamma \cdot C_e^{-1} \]

The subscript “\(m\)” means that the value of the physical quantities in the resulting equation refers to the mechanical characteristic during operation of the electric drive at the lower limit of the range; the subscript “\(n\)” means that \(M_c = M_n\); \(K'_{\text{ppu}}\) and \(K'_{\text{mu}}\) are the static gains of the amplifiers at the lower limit of the speed-control range.

Equating the indicated quantities to the current values at \(M_c = M_n\), we obtain the relationship between \(m\) and \(n\) for any speed setting \(n_n\), with excitation flux \(\phi = \Phi_{\text{nom}}\):

\[ m = \frac{100}{K'} \left( \frac{n_0 - 0.3 I_{yn} \cdot R_e \cdot C_e^{-1}}{n_n} - 1 \right) = f(n_n). \tag{14} \]

Here, in order to ensure constancy of the static accuracy over the range, one may take \(K' = K_{\text{ppu}} K_{\text{mu}} \gamma C_e^{-1} \Delta\% = \text{Const}\), while \(R_e\), in the general case, is a complex function of the speed \(n\). Assuming, for simplification of the analysis, that \(R_e = R_{er} = \text{Const}\), it may be considered that, in order to ensure \(\Delta\% = \text{Const}\) over the range, the relationship between \(m\) and \(n\) must be implemented according to a decreasing, approximately hyperbolic function of speed.

It should be noted that expression (14) is valid in the linear approximation and can be used to analyze the effect of \(m\) on the stiffness of the mechanical characteristics over the range. In an actual drive, \(K_{\text{ppu}}\), \(K_{\text{mu}}\), and \(R_e\) depend on speed; therefore \(m\) is a complex function of the electric-motor speed, and the correct selection of \(m\) over the range can be carried out only experimentally.

Figure 3 presents the static characteristics of the electric drive, obtained experimentally. It follows from them that, in the specified range, the use of a regulator with a variable gain coefficient

Fig. 3.

ensures approximate constancy of the static error as a function of the speed setting; at the same time, the stiffness of the mechanical characteristics, in accordance with the adjustment, is \(7 \div 8\%\) when the load current \(I_{\mathrm{я}}\) varies from 2 to 16 A, while the current limiting ensures a stall current that does not exceed the permissible value and, in accordance with the setting, lies within 24 A.

Figure 4 shows oscillograms \(n=f(t)\) and \(I_{\mathrm{я}}=f(t)\), recorded for speeds of 15, 30, and 3000 rpm:

a) for the case of starting without load; b) for application and removal of load, with the motor operating at speeds of 15 and 30 rpm; c) for switching from one speed to another.

Here:

\[ +\sigma^{\%}=\frac{n_{\max}-n_{\mathrm{н}}}{n_{\mathrm{н}}}\cdot 100\%, \qquad -\sigma^{\%}=\frac{n_{\min}-n_{\mathrm{н}}}{n_{\mathrm{н}}}\cdot 100\% . \]

— respectively, positive and negative overshoot,

\(\Delta \%\)—static accuracy, \(U_y\)—input signal,
\(M_c\)—static torque.

It can be seen from the oscillograms that the greatest speed sag, the transient-process time, the amount of overshoot, and the relative nonuniformity of the rotational speed caused by rotational and tooth torque pulsations are observed at the lower limit of the range \((D = 200:1)\), at \(n = 15\) rpm, which is characteristic of deeply regulated—

Fig. 4a.

Fig. 46, c

drives with magnetic amplifiers. However, as experiments in metal machining have shown, this does not exceed the requirements imposed on the feed mechanisms of the slides of heavy vertical lathes, for which the drive under consideration was developed, since under the conditions of the technological process the expected load disturbance is specified within the limits of \((15 \div 25)\%\) of the rated torque. Therefore, on the whole, the obtained indices may be regarded as quite satisfactory. The operation of current limiting is visible on the oscillogram of motor starting to the speed \(n = 3000\) rpm \((t_{\text{to}} = 0.6\ \text{s})\). Stabilization of the drive is reliably ensured over the entire range. The static error, in accordance with the setting, is \(\Delta \% = 7 \div 8\%\).

Fig. 5.

Figures 5a and b show, as obtained from processing oscillograms, the dependences \(\sigma = f(n)\) of the overshoot of the ACS in transients under the control action, and the curves \(\delta = f(n)\) of the index of nonuniformity of the rotational speed of the actuator motor over the range:

\[ \delta\%=\frac{n_{\max}-n_{\min}}{n_{\max}+n_{\min}}\cdot 100\%. \]

For comparison, the same figure also gives, in dashed lines, analogous dependences characteristic of the known EMU-D system [3] with an intermediate electronic amplifier and an actuator motor of the PN series.

The developed electric drive based on the MUS-D system with a class-D PPU is somewhat superior to the aforementioned EMU-D system with a PN motor in the quality of the transient process during drive starting, but is inferior to it in the magnitude of the nonuniformity of the rotational speed. The nonuniformity can be reduced by replacing the G-series motor with a PBST motor designed for operation under machine-building production conditions. The use of a class-D PPU made it possible to reduce the constant \(T_{\text{mu}}\) by redistributing the gain coefficients \(K_{\text{pur}}\) and \(K'_{\text{pur}}\) according to the power of the main ACS channel, while the reduction of the speed sag is achieved by applying nonlinear correction, which is discussed in detail in other papers by the authors and in [9].

Testing of the developed electric drive under the conditions of the UZTS plant on the feed mechanism of the slides of a model 1580 vertical lathe showed positive results.

Conclusions

1. The studies carried out have shown the possibility of creating a reliable electric drive with a 200:1 range, based on the MUS-D system with a class-D PPU, using standard MUS units of the UMZPT type or the serial PMU-9M-1 drive, the P-41 motor, and the TMG-30P tachogenerator, as applied to the feed mechanism of the slides of heavy carousel lathes, with satisfactory static and dynamic characteristics that are in no respect inferior to those of the installed EMU-D drive with a PN motor on the model 1580 carousel lathe, while having certain advantages inherent in drives with static current converters.

2. The use of a high-impedance speed setting unit based on the PP36 and, in connection with this, a low-power power supply reduced the dimensions and cost of the device.

3. Regulation of the gain factor as a function of the speed setting by changing the resistance of the input circuit of the PPU through the introduction of additional resistances into the potentiometer sections (by changing \(m\)) ensures approximate constancy of the “static error” over the range and contributes to improving the drive dynamics.

4. The adopted correction variant provides effective stabilization over the range, is characterized by a low level of interference, and has sufficient operating reliability.

5. The use of a class-D PPU made it possible, by redistributing the power gain factors \(K_{\text{ppur}}\) and \(K_{\text{mur}}\) of the main SAR channel, to reduce the time constant of the MUS when it operates in combination with the PPU, and to increase the speed of response of the MUS-D drive. Under these conditions, taking into account the transition to a MUS with a current control winding and PWM with \(f_c = 300\ \text{Hz}\), the constant can practically be reduced to zero and, with appropriate stabilization, the MUS-D converter will, in terms of speed of response, practically not differ from the BMU while having a simpler circuit solution.

6. The circuit solution of the PPU reliably protects it against overvoltages arising in the control windings during switching of the MUS power circuit; the most probable faults—an open input circuit, failure of the output triode (breakdown of the collector junction)—lead only to the initial zero output of the PPU, locking of the MUS, and stopping of the motor.

7. By applying nonlinear correction, if necessary, the speed drop and overshoot during load shedding can be significantly reduced.

8. Expansion of the speed-control range, with the specified stiffness of the mechanical characteristics and with dynamic indicators meeting the requirements, can be expected when using a PBST motor and a more advanced tachogenerator.

Literature

1. M. A. Rozenblat. Magnetic Amplifiers. Sovetskoe Radio Publishing House, vol. II, 1960.

2. A. Ya. Petrovskii, Ya. B. Rozman. Adjustable Electric Drive with Magnetic Amplifiers. Energiya Publishing House, 1964.

3. M. E. Verkholat. The present state of the electric drive of metal-cutting machine tools. Leningrad, LDNTP, Modern Automated Electric Drive, issue 4, 1963.

4. A. A. Sirotin. Automatic Control of Electric Drives. Gosenergoizdat, Moscow—Leningrad, 1958.

1. M. A. Borovikov. Dynamic processes in an electric drive based on the single-phase magnetic amplifier–motor system. Izvestiya VUZ, “Electromechanics,” 1965.

2. V. A. Sevastyanov, A. I. Esin, G. S. Sidorov. Analysis of circuit solutions for a PMU drive using magnetic amplifiers of a unified series. Electrical Machines and Automated Electric Drive. Publishing House of the Leningrad Polytechnic Institute, 1963.

3. V. A. Sevastyanov, A. P. Ineshin. On improving the static accuracy and stabilization of controlled electric drives, MUS-D. Scientific Notes of the Ulyanovsk State Pedagogical Institute. Vol. 20, issue 4, 1965.

4. V. A. Sevastyanov, A. I. Esin, G. S. Sidorov. On expanding the range of speed control of an electric motor in drives with magnetic amplifiers. Scientific Notes of the Ulyanovsk State Pedagogical Institute named after I. N. Ulyanov. Vol. 19, issue 2, 1964.

5. A. P. Ineshin, V. A. Sevastyanov. On increasing the response speed in systems with magnetic amplifiers. Scientific Notes of the Ulyanovsk State Pedagogical Institute. Vol. 20, issue 4, 1965.