ON THE MEASUREMENT OF THE FRICTION TORQUE OF INSTRUMENT BALL BEARINGS

E. F. KOLOTII, I. G. LESHCHENKO

(Presented by the scientific seminar of the Faculty of Automation and Computer Engineering)

This paper presents considerations on the further development and application of electrical methods for measuring the friction torques of ball bearings, which, in our opinion, are the most promising for the purpose of creating integrated installations for investigating dynamic and static friction torques and automatic machines for sorting bearings into serviceable and defective ones according to the entire program provided for by the technical specifications.

At enterprises of the bearing industry, under conditions of mass production, quality control of bearings according to their principal characteristic—the friction torque—is carried out selectively, which is connected, first of all, with the lack of sufficiently productive instruments. In recent years, at individual enterprises a number of instruments have been developed and introduced for measuring the dynamic friction torque and the starting torque (static friction torque), but these instruments and devices, for the most part of the mechanical type, measure only one parameter; to measure other parameters or repeat measurements, corresponding manual preparation of the apparatus is required. The semiautomatic instruments developed at bearing plants 2GPZ and 4GPZ, using photoelectric transducers for monitoring the dynamic friction torque by the running-out method and the shortened running-out method, are the beginning of the automation of bearing inspection only with respect to dynamic friction torque, with a productivity of about 250 bearings per hour.

The need for 100-percent inspection of bearings under production conditions is dictated not only by the fact that it will make it possible to raise significantly the technological level of production, especially of instrument bearings, for which the permissible friction torques amount to tenths and hundredths of gram-centimeters, but also by the fact that modern precision instrument engineering requires bearings of very high quality.

An analysis of the methods and instruments used [1], despite their apparent simplicity, shows that from the standpoint of increasing the productivity of complete testing (with respect to friction torque and starting torque) of bearings, the accuracy of measurements, and service characteristics, almost all these methods have little promise. As it seems to us, with the present state of electrical measuring technology, not all possibilities for monitoring such mass-produced goods in the national economy have by any means been exhausted.

To create automatic instruments for the complete testing of bearings, it is necessary to combine in a single installation the monitoring of the dynamic friction torque at specified speeds and of the starting torque at ten azimuths, as provided by the technical specifications for instrument ball bearings, with the issuance of test-result information in a form convenient for an inspector working on semiautomatic machines and for fully automatic sorting on automatic machines.

On the basis of existing pneumatic-type instruments, the DMP-2 pendulum type, the MDSh-1 with a water manometer, the MKT-1 based on the acceleration method, the S-24 based on the rundown method, and the semiautomatic instruments 2GPZ and 4GPZ, it is not possible to accomplish the stated task without very great expense. Moreover, for investigating dynamic friction torques from the starting torque up to maximum speeds amounting to tens of thousands of revolutions per minute, there are practically no test stands at all, except for one installation developed at the Institute of Mechanical Engineering of the Academy of Sciences of the USSR and intended for testing a pair of instrument bearings. Incidentally, this installation is complex and inefficient, since it requires a series of measurements of the angular velocities of the flywheel with oscillograph recording, construction of graphs of angular velocity versus time, and subsequent graphical differentiation, as a result of which the dynamic friction torque is determined for each time interval [2].

The most promising instruments for automating quality control of instrument bearings and for investigating their friction torques as a function of rotational speed are compensation-type instruments, based on electrical methods of measurement using various types of transducers and, in particular, rotating-type transducers.

Using the well-known advantages of electrical methods for measuring nonelectrical quantities and, in particular, the method of measuring bearing friction torques with rotating transducers [3, 4], the Department of Electrical Measuring Engineering has obtained the necessary positive results, which make it possible to assert the very great promise of these methods for comprehensive automation of bearing inspection by friction torques.

Below, only the compensation method for measuring dynamic friction torque using an electromagnetic transducer is set forth in general form; other methods (the induction method for measuring friction torque and starting torque, the electromagnetic method for measuring starting torque, and the time method for measuring friction torque) are described in subsequent articles of this same collection.

For the purpose of measuring the dynamic friction torque of instrument ball bearings by the compensation method with the axis vertical, it is proposed to use an electromagnetic rotating transducer. The operating principle of the device with an electromagnetic rotating transducer is as follows. When shaft 2 (Fig. 1) rotates at a previously known constant speed \(\omega\), the outer ring of the bearing under test 1, together with flywheel 3, which creates an axial load, is carried along in the direction of rotation of the shaft by the friction torque of the bearing. The friction torque is compensated by the torque developed by electromagnet Э (Fig. 2), applied to the moving part through core 5. The magnitude of the current through the electromagnet is automatically established by a device consisting of a bridge circuit with a pho-

resistances, an illuminating device (not shown in the figure), the shutter 4 that interrupts the light beam, and an amplifier. When the bridge is in the balanced state, the voltage at the amplifier input is zero and no current flows through electromagnet E. If a voltage appears at the bridge output, then current flows through the electromagnet, which tends to turn the moving part so as to return the bridge to the equilibrium position. When shaft 2 rotates, the moving part of the device assumes a new equilibrium position when \(M_{\mathrm{tr}} = M_{\mathrm{e}}\); in this case there is an unbalance voltage at the bridge output.

Fig. 1. Diagram of the moving part.

Fig. 2. On the principle of operation of the instrument with an electromagnetic inverse transducer.

The structural diagram of the device with an electromagnetic inverse transducer is shown in Fig. 3.

Two oppositely directed torques act on the moving part of the instrument: the friction torque \(M_{\mathrm{in}}\) and the torque of the electromagnet \(M_{\beta}\). Under the action of the difference of the torques \(\Delta M\), the moving part of the instrument turns through a certain angle \(\alpha\) relative to its initial position. This causes, first, a change in the size of the gap \((\delta)\) between the core of the electromagnet and the moving core 5 (Fig. 2) and, second, a change in the current \(I_{\mathrm{out}}\) through the electromagnet. Both these quantities \((\delta\) and \(I_{\mathrm{out}})\) are functionally related to the output quantity of the electromagnet \(M_{\beta}\). In the equilibrium state of the moving part \(M_{\mathrm{in}} = M_{\beta}\) and \(\Delta M = 0\), but \(\alpha \ne 0\). By measuring in this case the current through the electromagnet, one can judge the friction torque of the ball bearing. Thus, ammeter \(A\) can be graduated directly in units of torque measurement.

Fig. 3. Structural-functional diagram of the instrument with an electromagnetic inverse transducer.

From the structural diagram of the instrument it is clearly seen that this device is a closed-loop automatic-control system; therefore, for solving questions of stability, speed of response, finding the transient characteristic and the transfer function of the device, one can use the mathematical apparatus that has been developed in depth in the theory of automatic control.

The presence in the instrument of additional feedback \((\alpha \to \delta \to M_\beta)\) leads to a reduction in the static accuracy of measurement in comparison with single-loop systems, since in this case the error in the transformation coefficients of the chain \(\alpha \to U_{\text{in}} \to I_{\text{out}}\) enters into the total error of the instrument.

The principal advantages of a device with an electromagnetic reversible transducer for measuring the dynamic friction torque of ball bearings, as compared with the instrument with a magnetoelectric reversible transducer described in the literature [3, 4, 5], are:

a) the possibility of measuring significantly larger friction torques;

b) the absence of current leads to the moving part and, as a consequence, the possibility of more rapid replacement of the bearings under test, which increases the productivity of the instrument.

In order to obtain the possibility of measuring the dynamic friction torque when the direction of rotation of the shaft is changed, one more electromagnetic transducer may be used, which would create a torque applied to the moving part and directed oppositely to the friction torque. In this case, at zero voltage at the output of the amplifier, the currents through both electromagnets must be equal to zero (or be identical). The polarity of the voltage at the amplifier input must control the change in the currents through the electromagnets in such a way that rotation of the moving part under the action of the tractive force of the electromagnets leads to a decrease in the unbalance voltage of the bridge with photoresistors.

A mock-up of the device with two electromagnetic transducers was manufactured and tested at the Department of Electrical Measuring Technology of the Tomsk Polytechnic Institute; the operation of the mock-up was satisfactory with deep damping of the moving part.

Thus, from the example of the application of an electromagnetic reversible transducer for measuring the dynamic friction torque of ball bearings, it is clearly seen that the possibilities of applying electrical methods for measuring the parameters of instrument ball bearings are far from exhausted, and that it is possible to create new, more accurate and more productive electrical instruments for measuring the friction torque of ball bearings.

REFERENCES

1. N. V. Bruevich, Z. M. Breitman, Yu. M. Reznikov. Technical Measurements in the Bearing Industry. Mashgiz, 1963.

2. R. M. Matveevskii. Investigation of Friction in Instrument Ball Bearings. Mashgiz, 1957.

3. A. M. Turichin. Electrical Measurements of Nonelectrical Quantities. GEI, 1959.

4. M. M. Fetisov. New Automatic Compensation Instruments for Measuring Nonelectrical Quantities. Parts I and II, LDNTI, 1960.

5. P. M. Pflier. Electrical Measurements of Mechanical Quantities. Mashgiz, 1948.