Vibratory Conveying of Parts

(experimental results)

G. V. Surkov

(Presented by the scientific seminar of the Departments of Gyroscopic Instruments and Devices and Mechanics)

The large number of factors influencing the process of vibratory conveying makes it difficult to determine the feed rate theoretically in a general form. Therefore, experimental data are widely used to establish the regularities in the variation of the velocity and for the subsequent correction of particular theoretical formulas.

Below are presented the results of processing experiments on the velocity of conveying parts on a vibratory conveyor with an electromagnetic drive, depending on the vibration parameters, the type of conveyed load, the coefficient of friction, and the surrounding medium.

The design of the experimental vibration setup with a rectilinear trough makes it possible to vary the vibration parameters over wide limits: the oscillation amplitude \(A\), the natural tuning frequency, the angle of inclination of the trough \(\alpha\), and the vibration angle \(\beta\) (the angle between the direction of oscillations and the direction of motion). The parts are conveyed over a technically smooth steel surface. In order to eliminate the influence of the moving masses on the oscillations of the conveyor itself, the mass of the load was taken to be ten times smaller than the mass of the trough. The traction electromagnet is single-stroke and is supplied from the industrial mains through a selenium rectifier. The natural frequency of the conveyor in air is \(56\ \text{Hz}\), and in water \(54\ \text{Hz}\). The magnitude of the double amplitude was measured with an MBS-1 microscope. The time required for a part to traverse a precisely defined path along the trough was recorded by a photoelectric time marker, which consists of two photoelectric relays and an electric stopwatch PV-52. The accuracy of measuring the path was \(\pm 0.5\ \text{mm}\), and of reading the time \(\pm 0.01\ \text{s}\).

In studying the process of vibratory conveying in a liquid, the dimensions of the basin were chosen large enough to eliminate the influence of the side walls of the basin and of the free surface of the liquid on the magnitude of the added mass during the unsteady motion of the parts [1].

As a result of an experiment, the time required for a part to traverse the control path was determined. Each experiment was repeated 10 times. Some experiments were repeated 15 times. Calculations showed that the mean velocity determined from 15 repetitions of an experiment is practically equal to the velocity obtained from 10 measurements. The discrepancy according to the results of 8 experiments is no more than 2%.

The vibratory conveying of piece products was investigated on cylindrical blanks of diameter \(30\ \text{mm}\), length \(90\ \text{mm}\), part No. 1

made of duralumin, part No. 2 of steel, and part No. 3 of lead. The influence of the mass of the load on the conveying speed was studied on a vibratory feeder with a hopper 400 mm in diameter [2]. Blanks 10 mm in diameter and 30 mm long were made: part No. 4 of duralumin, No. 5 of steel.

Fig. 1. Dependence of speed on vibration angle.

According to the experimental data, Fig. 1 plots the dependence of the vibratory-conveying speed on the vibration angle \(\beta\) at a trough inclination angle \(\alpha = 0^\circ\) and a trough oscillation amplitude \(A = 0.75\) mm. Processing of the experimental results by formula (14) [3] at \(\eta = 1\) (dash-dot line) shows good qualitative agreement between theory and experiment. The quantitative agreement should be taken into account by the correction coefficient \(\eta\).

The values of the correction coefficients, determined from a comparison of the experimental data and the theory at a discrepancy of about 5%, are: \(\eta_1 = 0.48\) for the motion of part 1, \(\eta_2 = 0.61\) for the motion of part 2 in air, and \(\eta_2' = 0.37\) for the motion of part 2 in water. In the zone

Fig. 2. Influence of the trough oscillation amplitude on speed.

with vibration-regime parameter \(\Gamma = 3.3 \times 4.6\) (in Fig. 1 the zone is hatched), the speed decreases noticeably, but to a somewhat lesser extent than the theory predicts.

It is seen from Fig. 1 that, at the dynamic coefficient \(K = 7.55\), there exists an optimum value of the vibration angle \(\beta \cong 22^\circ\). Theoretically [3], for this value of the dynamic coefficient, the optimum value of the angle is \(\beta \cong 21^\circ\); that is, the theoretical conclusions for determining the optimum vibration angle in this case agree well with experiment.

Figure 2 gives the dependence of the velocity on the amplitude of oscillation of the trough at \(\alpha = 0^\circ\), \(\beta = 20^\circ\). All the curves are plotted from the formula for the above-indicated values of the correction coefficients, except for the velocity plot for part 1 in water, which has a considerable discrepancy (up to 30%) from the theoretical data.

Vibratory conveying of parts in a liquid is characterized by certain special features. The experiment shows that the drop in the velocity of motion in water is greater the smaller the specific weight of the parts and the higher the value of the vibration parameter \(\Gamma\) (Figs. 1, 2). It may be concluded that the speed of vibratory conveying is markedly affected by hydrodynamic resistance. The influence of the resistance force and of other processes associated with disturbance of the surrounding volume of liquid (suction, added mass of the liquid) on the process of motion increases as the specific weight of the part decreases and as the flight of the part during a cycle of motion increases.

Fig. 3. Influence of the reduced coefficient of friction on velocity.

The influence of the coefficient of friction on the feed velocity was investigated on parts 1 and 2. As the experiment showed, the conveying velocity is influenced to a greater extent not only by the physicomechanical characteristics of the part and the trough, but also by the shape, and even by the orientation of the part on the trough. Therefore, in the experiment the reduced coefficient of friction was varied by changing the angle of the helical track of the trough, equal to \(2\Theta^\circ\). The conveying velocity at \(\alpha = 0^\circ\), \(\beta = 20^\circ\), \(\Gamma = 2.58\) (Fig. 3) may be considered, under the experimental conditions, to be independent of the coefficient of friction, which is in complete agreement with the theory [3].

Fig. 4. Average speed of delivery of parts from the hopper as a function of the mass flow of the load.

In the case of transporting bulk loads, the feed rate is affected by the hopper track and other factors that reduce the average speed at which parts exit the vibratory feeder. Figure 4 gives the results of the experiment. It may be assumed that the dependence of the average exit speed on the graph in logarithmic coordinates is described by a straight line, that is, \(v_N = vN^{-m}\), where \(v\) is the transport speed of a single part and \(m\) is the experimental exponent. In the present case \(m = 0.038\). The feed rate of a mixture of parts characterized by different speeds of motion can be determined as the arithmetic mean value (Fig. 4).

According to the experiment and the results of processing, it has been established that the speed of vibratory conveying of individual parts in the regime with tossing agrees qualitatively with the theoretical data in the case where hydrodynamic resistances do not exert a substantial influence on the free motion of the parts. The quantitative relationship is established by correction coefficients.

For a more complete disclosure of the regularity of variation in the speed of vibratory conveying, it is necessary to take into account the physicomechanical properties of the parts, the working surface of the conveyor trough, and the resistance of the surrounding medium.

REFERENCES

1. I. S. Riman, R. L. Kreps. Added masses of bodies of various shapes. Proceedings of TsAGI, No. 635, Oborongiz, 1947.

2. Yu. Ya. Kovylin, G. V. Surkov, V. N. Titov. Stanki i instrument, No. 5, pp. 25–27, Mashinostroenie, 1964.

3. G. V. Surkov. Problems in the theory of vibratory conveying. Present collection.