SETUP FOR THE EXPERIMENTAL STUDY OF ELECTROMAGNETIC VIBRATORS AND SOME EXPERIMENTAL RESULTS

Yu. Ya. Kovyliin

To verify the results of theoretical studies of electromagnetic vibrators presented in works [1—5], we created an experimental setup. Its general form is shown in Fig. 1.

A standard audio generator 1 of type GZ-2, serving as a source of alternating electrical voltage of calibrated frequencies, is used as the reference generator. Its output is connected to the input of an original high-quality amplifier 2 with an undistorted power of up to 0.5 kW.

The output stage of the amplifier is made according to a push-pull circuit. In each arm there operate two tubes of type GM-70. The circuits of the control and screen grids of the output tubes contain switchable \(RC\) networks. This makes it possible, over a wide range, to regulate the shape of the electrical-voltage curve at the amplifier output, which can be observed visually on the screen of a cathode-ray oscilloscope 3, for example, of type SI-1.

For supplying the anodes and screen grids of the output tubes, two universal power supplies 4 of type UIP-1 are used, with their outputs connected in series. With a load of up to 0.2 kW, operation from a single UIP-1 is possible. The filament circuits of all tubes of the power amplifier are fed with direct current from an autonomous stabilized power supply placed on the common chassis with the amplifier. This makes it possible to eliminate the 50 Hz component at the amplifier output, which interferes when testing vibrators whose elastic-system natural frequency is about 50 Hz (vibrators with such tuning are at present widely used in instrument- and machine-building).

The magnitude of the filament voltage is monitored by a voltmeter installed in the middle part of the amplifier’s front panel. To the left and right of it, on the same panel, are installed a voltmeter and an ammeter used for rough monitoring of the effective values of the voltage at the amplifier output terminals and of the current consumed by the windings of the electromagnetic vibrator under test. More accurate measurement of these electrical quantities is carried out with the aid of electrostatic multi-range voltmeter and ammeter 5 and 6.

The experimental electromagnetic vibrator 7 is shown separately in Fig. 2. It consists of an electromagnet 1 made on an Ш-shaped core assembled from varnished sheets of grade USh-32 electrical steel. The magnet coil is replaceable. This makes it possible to carry out experiments with different numbers of turns and different cross-sectional diameters of the winding wire.

Fig. 1.

Fig. 2.

The electromagnet is suspended from a steel measuring beam 2, the ends of which are clamped between the faces of nuts 3. These same nuts are used to adjust the magnitude of the air gap between the electromagnet and the armature 4. The latter consists of a set of rectangular varnished plates, also made of the same steel as the core of the electromagnet, and fixed in a frame rigidly connected to two leaf springs 5. The ends of these springs are clamped in vertical posts mounted on the same massive cast-iron base 6 as the columns 7 carrying the measuring beam 2. The armature is provided with an adjustable hydraulic damper, which makes it possible to simulate the forces of inelastic resistance in the vibrator and ensures high stability of the amplitudes of armature oscillations in near-resonance operating modes of the vibrator.

The thickness of the leaf springs 5 and the distance between the posts can be varied. In this way, different values of the stiffness of the elastic links can be obtained. The columns 7 are made of nonmagnetic stainless steel so that no bypass paths for the magnetic flux arise in the vibrator. In the working position, the base 6 rests on soft rubber shock absorbers.

The instantaneous forces developed by the electromagnetic vibrator during its operation are measured by means of wire strain gauges glued in the form of a half-bridge onto the measuring beam 2. In series with the working strain gauges, one additional strain gauge is connected, glued to the compensating beam 8. The parameters of the latter are selected so as to suppress disturbances caused by kinematic excitation of the measuring beam from vibrations of the base 6.

The high-frequency components of the vibration of the measuring beam, arising from its resonances at higher harmonics of the disturbing force, are suppressed by a Π-shaped \(LC\) filter provided in the force-recording channel.

Parasitic resonances of the compensating beam are suppressed by damper 9, made in the form of a plastic plate.

Another half-bridge of wire strain gauges is glued to the upper leaf spring 5 and serves to record the vibrational displacement of the armature by measuring the mechanical stresses in these springs that arise when the armature deviates from the equilibrium position.

Both half-bridges are connected to the input of strain-gauge station 8 (Fig. 1), type 8ANCH-7m. The signal amplified by the station is recorded on film by loop oscillograph 9, type N-102. On the same film are simultaneously recorded the instantaneous values of the current consumed by the electromagnet winding and of the electrical voltage at the output terminals of the power amplifier, as well as time marks (500 Hz).

The amplitude of oscillations of the vibrator armature relative to the electromagnet can be determined from the width of the luminous strip visible against the background of the reference scale in the field of view of instrumental microscope 10, type MIR-1, with a division value of 0.05 mm. For this purpose, a Plexiglas plate is attached to the armature; on the lateral surface facing the microscope there is a very fine scribed line, brightly luminous under top illumination.

A clock-type indicator 11 is used to measure the static displacement of the armature when calibrating the recording apparatus. During operation of the vibrator it is removed.

The amplifiers of the 8ANCH-7m strain-gauge station were fitted with new radio tubes. After this, the strain-gauge station was “run in” simultaneously with the vibrator under test for 24 hours. This proved sufficient so that subsequently, when carrying out experimen-

the zeros did not drift during 6 hours of continuous operation of the vibrator. Each day, after being switched on, the station warmed up for 2 hours. Thus, from the standpoint of operating stability, the 8ANCh-7m strain-gauge station (in the aviation version) deserves the highest rating. Conversely, the lowest rating is deserved by the N-102 loop oscillograph of the Leningrad “Vibrator” plant, which, though new, proved completely unsuitable for serious use and required a large amount of repair and adjustment work before reaching its rated parameters.

Before the tests were carried out, the setup was carefully calibrated. By means of weights placed on a special stirrup suspended from the electromagnet, the force-recording channel was calibrated in divisions of the scale of the N-102 oscillograph. By tightening a nut screwed onto the rod of the hydraulic damper, the armature was displaced. The amount of displacement was determined from indicator 11 and in divisions of the scale of the oscillograph.

Fig. 3a

Fig. 3b

The values of the electrical voltage at the vibrator terminals (before the rectifier) were estimated in divisions of the same scale.

During calibration it was established that the adopted dimensions of the leaf springs ensure the linearity of the elastic force within the limits of armature oscillations with an amplitude of up to 2.5 mm; when this value is exceeded, only a very slight nonlinearity is observed, characteristic of “stiff” springs. This makes it possible to compare the experimental results with the theoretical studies of vibrators mentioned at the beginning of the article, which were carried out under the assumption of linearity of the characteristics of the mechanical elastic couplings of the vibrator’s oscillatory system.

The experiments confirmed the acceptability of all the basic assumptions underlying these works.

In Figs. 3a and 3b, by way of example, two oscillograms of processes in an electromagnetic vibrator are shown, recorded experimentally, together with calculated curves obtained by the method set forth in [1]. The parameters of the vibrator and the settings of its oscillatory system that influence the processes are written in the margins of the calculated oscillograms (in the designations of [1]).

Article [2] gives an example of an oscillogram for the third (near-resonance) operating mode, also obtained on the described setup, and confirming the acceptability of the calculation method presented there for near-resonance modes of the vibrator.

The good agreement of the experimental and calculated curves for all the main internal processes occurring in the vibrator also ensures good agreement of its external characteristics. The conclusions of [1], [2], and [3] are confirmed: namely, that when the active resistance in the supply circuit of the magnet winding is small, with the winding fed from a sinusoidal voltage source, both in the absence of a diode in this circuit and in its presence, the amplitude–frequency curves of the vibrator have the form characteristic of an ordinary linear oscillatory system with one degree of freedom.

The results of [4] and [5] are also confirmed: when the winding is supplied from a current source (the case of a large active resistance in the supply circuit of the magnet winding), the maxima of the amplitude–frequency curves shift to the left, and the more so the larger the amplitude of the armature oscillation. With below-resonance tuning of the oscillatory system, in this case a region of unstable oscillations appears.

The theoretical data on the energy consumption of the vibrator, presented in article [2], are well confirmed.

In creating the experimental setup described, the author received substantial assistance from student L. N. Drozdov.

REFERENCES

1. Yu. Ya. Kovylin, “Toward the foundations of the theory of single-cycle electromagnetic vibrators with a half-wave rectifier” (this collection).

2. Yu. Ya. Kovylin, S. I. Basov. Interaction of an electromagnetic vibrator with a sinusoidal voltage source. Physicotechnical Problems in the Processing of Minerals. Novosibirsk, Publishing House of the Siberian Branch of the USSR Academy of Sciences, No. 4, 1965.

3. Yu. Ya. Kovylin. Oscillations of an electromagnetic vibrator operating from a sinusoidal voltage source without a rectifier. Tomsk, Proceedings of TPI, vol. 139, 1965.

4. Yu. Ya. Kovylin. On the question of the nature of the amplitude–frequency curves of an electromagnetic vibrator operating from a source of pulsating current. Tomsk, Proceedings of TPI, vol. 139, 1965.

5. Yu. Ya. Kovylin. Elastic oscillations of a ferromagnetic body in the field of one magnet. Proceedings of Higher Educational Institutions of the USSR, Electromechanics. Novocherkassk, No. 5, 1968.