Proceedings of the Tomsk Order of the Red Banner of Labor Polytechnic Institute named after S. M. Kirov

Vol. 211 1970

ROTOR POSITION SENSOR AND SEMICONDUCTOR COMMUTATOR FOR BRUSHLESS DC MOTORS

L. Ya. ZINNER, V. P. KOROTAEV, A. I. SKOROSPEШKIN

(Submitted by the scientific seminar of the departments of electrical machines and general electrical engineering)

Because DC electric machines have a number of substantial shortcomings caused by the presence of a commutator-brush assembly, at present increasing attention is being devoted to the development of brushless electric machines.

Fig. 1. Rotor position sensor.

The principal units of brushless DC motors are the controlled semiconductor commutator (UPK) and the rotor position sensor.

Existing inductive sensors of the choke and transformer types have significant disadvantages: a low steepness of the signal fronts and a small ratio of output signal to input signal [1]. In connection with this, the authors of the present article developed a rotor position sensor free of these disadvantages.

The operation of the sensor is based on the eddy-current method. The design of the position sensor is shown in Fig. 1. The eddy-current sensor is a U-shaped composite core made of 1000 nn ferrite. Its two poles 1, 2, with grooves previously machined for the signal winding $W'_c$,

glued together through shielding copper foil so that the minimum distance between them is 0.015–0.02 mm. The poles are connected by a carefully ground yoke 3 with the excitation winding \(W_{\mathrm{v}}\) placed on it. The opposite end of the pole block is ground into a wedge. The magnetic core of the sensor, with its windings, is placed in a copper shield (the sensor housing). Only the ends of the poles of the ferrite core emerge at the surface, forming the working surface of the sensor, with dimensions \(3 \times 0.3\) mm. The shield concentrates the electromagnetic field of the sensor in the half-space adjoining the ends of the core, directing the lateral leakage fluxes into this region.

When a conducting nonmagnetic material (the sensor armature) approaches the end surfaces of the sensor, the electromagnetic field of the eddy currents excited in it, acting on the electromagnetic field of the sensor, changes the electrical parameters of the signal winding \(W_{\mathrm{s}}\). The introduced inductance, having a negative sign, increases the resonant frequency of the signal circuit, while the introduced active loss resistance reduces its quality factor. The separate arrangement of the windings (the excitation winding in the middle part, the signal winding at the ends of the core), when the sensor approaches the armature, leads to a decrease in the coupling coefficient between the windings. Each of these factors separately causes a change (decrease) in the voltage at the sensor output when the distance between it and the conducting surface of the armature changes (decreases).

The armature of the rotor-position sensor is a copper disk that is fastened to the shaft inside the housing of the electrical machine. It has segments in the form of steps. The height difference between adjacent segments is 0.1 mm. The number of segments is equal to the number of commutated sections of the motor. The ratio of the duration of the alternating-current pulses taken from the signal winding of the sensor to their repetition period is determined by the angular dimension “\(\beta\)” of the signal sector. When the number of commutated sections of the electrical machine changes, the magnitude of the signal sector also changes and, consequently, so does the relationship between the pulse duration and its repetition period. In the case under consideration the motor has the number of commutated sections \(a_{\mathrm{s}} = 3\); the signal sector of the sensor armature is equal to \(120^\circ\), and then the pulse duration is equal to \(1/3\) of the period.

The design arrangement of the sensor is such that the steepness of the edges of the output pulse depends on the height of the working surface of the sensor \(h = 0.3\) mm and on the radius of the sensor armature. Owing to the small height of the working surface of the sensor, the latter has high resolving power, as a result of which the output-voltage pulse has a very short edge duration, and the ratio of the edge duration of the pulse (the total rise and fall time of the signal) to their repetition period is 0.006–0.008, i.e., an order of magnitude less than in existing inductive sensors [1]. Consequently, the switching time of the commutator transistors is sharply reduced, the power losses dissipated in the transistors are decreased, and the economy of the commutator is increased. The sensor can control a motor of any power, is well matched to the commutator circuit, and has the same operating-temperature range as the motor itself. It can operate reliably at high humidity, in an aggressive medium, under radioactive irradiation, and in a highly rarefied atmosphere.

The sensor is supplied from a rectangular-pulse voltage generator with frequency \(f = 50\) kHz and voltage 7 V, which at the same time makes it possible to obtain additional galvanically isolated sources of reference blocking voltages and voltages for feeding the negative-feedback (NFB) circuit.

The commutator circuit shown in Fig. 2 makes it possible to manage with a single rotor-position sensor for controlling a three-phase motor by means of

introducing negative feedback and different levels of the reference locking voltage of the transistors.

In order to increase the sensitivity of the position sensor, a compensating sensor (CS), identical to the first, is introduced into the commutator circuit; it compensates the initial output voltage of the position sensor at a quite definite gap. The excitation windings of both sensors are connected in series and in aiding polarity to the generator load, while the signal windings are connected in series and in opposition to the rectifier bridge. In this case the position sensor is installed at a distance of 0.1–0.15 mm from the armature surface relative to the segment with the greatest radius.

Fig. 2. Block diagram of a semiconductor commutator.

After rectification, the control voltage from the sensor switches on the trigger device 1, 2, or 3 of the corresponding commutator channel, depending on which of the armature segments is at that moment under the position sensor. The trigger device opens its transistor switch 4, 5, or 6, which connects the corresponding section of the electric machine into the supply circuit.

As it rotates and simultaneously turns the sensor armature, the rotor brings the next armature segment under the sensor and thereby changes the gap between the armature and the sensor itself; as a result, another control signal, differing in magnitude from the first, is delivered from the position sensor to the trigger devices. The commutator channel whose trigger device is adjusted to this signal magnitude operates.

To prevent the channels adjusted to a smaller signal from operating in response to the largest control signal, negative feedback based on an emitter-coupled trigger [3] is introduced into the commutator circuit. The trigger is assembled on transistors of the n-p-n type and is powered from the square-pulse voltage generator through a rectifier.

When the transistor switch of the corresponding channel, adjusted to the largest control signal, is switched on, the trigger operates, and almost the entire trigger supply voltage is applied to the bases of the transistors of the trigger devices of the remaining channels and locks them. Only one channel operates until the next control signal appears.

The introduction of a compensating sensor into the commutator circuit makes it possible to reverse the motor in a simple manner. For this purpose, compensation of the initial voltage of the position sensor is performed when an armature segment with a different radius is located under the position sensor. In

In this case the order of alternation of the control signals from the position sensor changes, and the motor rotor will begin to rotate in the opposite direction.

Thus, the developed eddy-current sensor, possessing high sensitivity, good resolving power, and relatively small dimensions, makes it possible to use it as a rotor-position sensor for a brushless DC motor. The commutator circuit makes it possible to manage with only a single rotor-position sensor for controlling a three-phase motor. In principle, it is possible to control, with one rotor-position sensor, a motor having any number of commutated sections. To do this, it is necessary to add the corresponding number of commutator channels, and the armature of the sensor must have a number of segments equal to the number of commutated sections.

REFERENCES

1. I. E. Ovchinnikov, N. I. Lebedev. Brushless DC Motors for Automatic Devices. Moscow—Leningrad, “Nauka,” 1966.

2. A. I. Bertinov, V. L. Lotoshkii. Brushless DC Electric Machines. Informstandartelektro, 1967.

3. I. P. Stepanenko. Fundamentals of the Theory of Transistors and Transistor Circuits. “Energiya,” 1967.

4. A. I. Proshin, L. Ya. Zinner, A. I. Skorospeshkin. “An Eddy-Current Sensor for Measuring the Profile of Collectors of Electrical Machines.” Izvestiya TPI, vol. 160, 1966.